Developing tools to help predict future risk for diabetes and heart attack

Future risk for people living with type 2 diabetes and those who have had a heart attack in the past can now be predicted as a result of research at Manchester.

Prof Evan Kontopantelis

Prof Evan Kontopantelis

Evan is Professor of Data Science and Health Services Research at The University of Manchester.

View academic profile

Type 2 diabetes

The problem

There are around 4.9 million people in the UK living with diabetes, a condition where the amount of glucose (sugar) in the blood is too high. Insulin allows glucose in our blood to enter our body's cells for energy. People with type 2 diabetes don't produce enough insulin or the insulin does not work properly, so the body can't absorb the glucose it needs and it stays in the blood.

Excess blood sugar can damage arteries and increases the risk of heart and circulatory problems. The NHS spends around £10 billion a year on diabetes, mainly because of complications such as heart attack, stroke, amputation and blindness.

The Diabetes Severity Score (DISSCO)

Our research has led to the development of a new tool for assessing the severity of type 2 diabetes, The Diabetes Severity Score (DISSCO) which could help personalise treatment to help prevent or delay complications and save NHS funds.

Our research, funded by NIHR, looked at anonymised health data from 2007 to 2017 and identified 34 indicators of the severity of type 2 diabetes. Other risk factors such as age, gender, ethnicity and socio-economic deprivation information were also considered in the assessment of the severity algorithm.

Better at predicting outcomes

Patients were assessed with DISSCO at diagnosis. Each DISSCO unit increase was found to be associated with a 14% higher risk of death and a 45% higher risk of hospitalisation due to cardiovascular complications.

Overall, DISSCO was found to have a higher predictive value than the current clinical practice of measuring glycated haemoglobin (HbA1c) levels in blood to assess the severity of the diabetes.

Next steps

Our computer algorithm could be developed into software for use in primary care. Having a score that can accurately assess who is most at risk could help to personalise treatment and target NHS resources where they are most needed.

Acute myocardial infarction (heart attack)

The problem

Several tools exist to predict the risk of future cardiovascular events generally. However, there is no tool to predict risk for people who have already had a heart attack.

In the UK, around 1.4 million people are living after a heart attack, meaning they are at higher risk of stroke, heart attack or death than people without cardiovascular disease. A tool to assess the future risk for these patients was needed. 

Studying the risk factors

Funded by Heart Research UK, our researchers used anonymised health data collected between 2006 and 2019 to create a tool to predict people's risk of having another cardiovascular event or heart attack within both 12 months and 5 years of a first heart attack.

The effects of other important risk factors such as existing health conditions, medications, body mass index (BMI), age, the amount people smoke and drink, and socio-economic deprivation information are also considered.

Promising early results

Initial results showed that patients with acute myocardial infarction who were older, males, current smokers, with diabetes, and living in deprived areas were more likely to die within 12 months and 5 years.

Next steps

Analysis is ongoing and we aim to publish our results in the next few months. Our hope is that the tool could help GPs to identify patients at higher risk of future heart disease events and death, so that appropriate medical treatment can be given and lifestyle changes can be made to reduce their risk.