04:24am Thursday 24 October 2019

Diabetes screening may not reduce deaths

In the first ever randomised controlled trial of diabetes screening, researchers found no evidence that testing people for the disease reduces the number of deaths from any cause, including those unrelated to diabetes, after ten years. The results, published today in The Lancet1, contradict several previous studies2, which suggested a widespread screening programme would be effective.

Dr Simon Griffin, from the Medical Research Council (MRC) Epidemiology Unit in Cambridge, said:

“Our study was the first robust evaluation of diabetes screening and the results suggest its effectiveness may have been overestimated. Based on our findings, screening is only likely to benefit the small minority of people living with undiagnosed diabetes and is unlikely to reduce deaths in the general population.

“While this doesn’t necessarily mean that screening for diabetes is invalid, it does suggest that tackling the diabetes epidemic is likely to require a more holistic approach using a combination of preventative strategies aimed at detecting and treating the disease early and reducing the risk of it developing in the first place.”

The study was funded by the Wellcome Trust, MRC, NHS and the National Institute for Health Research.

Assessment for diabetes is currently included in the NHS Health Check programme offered to all those aged between 40-74 years in the UK. The disease has many attributes that make it a good candidate for population screening: it is relatively common and prevalence is rising; it takes a long time to develop so there is a large window for early detection; and early treatment can minimise the risk of complications developing over time.

There has been some uncertainty around the benefits of population-based screening for type 2 diabetes. While several studies have suggested that a screening programme would be effective, these were largely based on computer simulations. This study set out to evaluate such screening by looking at one measure of effectiveness – whether it reduced deaths.

The researchers studied more than 20,000 people aged 40-69 years registered at 32 GP practices in the east of England, which were randomly allocated to screening or no screening (control clinics). In the screening clinics, patients at high risk3 of developing diabetes were tested for the condition and of these 2.9 per cent were diagnosed with the disease.

All participants (with and without diabetes) were then followed-up for an average of 9.6 years. After this time the researchers found no significant difference in the total number of deaths in the screened group, compared with the control group. There were also no differences in the rates of deaths attributed to cardiovascular conditions, cancer or diabetes-related causes. Cancer was the most common cause of death.

The study looked at only one geographical region, which had a largely Caucasian population and a higher than average socio-economic status. It is possible that the results would be different in other areas of the UK and in other countries.

Professor David Lomas, Chair of the MRC’s Population and Systems Medicine Board, said:

“Diabetes poses one of the biggest challenges to our health service. Already around three million people in the UK have the condition and the prevalence is on the rise. It’s vitally important that we devise effective strategies to not only treat the effects of diabetes, but also to prevent people from developing it in the first place. Large population studies like this one provide the robust evidence needed to inform such strategies.”

Notes to editors

The paper, entitled ‘Effect of screening for type 2 diabetes on population mortality over ten years: the ADDITION-Cambridge cluster-randomised controlled trial’ by Simmons et al, and an accompanying comment from Michael Engelgau of the National Center for Chronic Disease Prevention and Health Promotion are published in The Lancet.


Modelling studies suggest that screening a population every five years would lead to a 26-40% reduction in diabetes-related mortality.


A risk score was determined based on age, sex (men are more likely to develop T2D than women), family history, BMI, smoking habits and whether they were being prescribed drugs for high blood pressure.


For more information, please contact:


Hannah Isom

Senior Press Officer, Medical Research Council

T: 0207 395 2345 (out of hours: 07818 428 297)

E: press.office@headoffice.mrc.ac.uk

Share on:

Health news