This site is intended for health professionals only

Health checks scheme fails to identify a third of patients at risk of diabetes

A third of patients at high risk of having or developing diabetes may be missed by the NHS Health Checks scheme, shows a new study raising questions over Department of Health plans to radically expand the programme across England.

The study showed that the scheme was systematically overlooking patients with high HbA1c levels, but with normal or low body weight, leading the researchers to recommend a change to the way it is run.

The findings come a week after the DH signalled its intention to radically expand the scheme by pledging to widen access to the health checks for all 40 to 74 year olds in its strategy for cardiovascular disease.

But GP experts say that the scheme is proving ‘underwhelming’ in the number of high-risk patients that it is throwing up, and is widening health inequalities.

The NHS Health Check scheme was set up in 2009 and was designed to improve the nation’s life expectancy and reduce health inequalities, by engaging with people about potentially modifiable risk factors and improving case-finding.

But it has suffered from low uptake, and a varied rollout by PCTs across the country.

This latest study, published in BMJ Open, comes from one of the most enthusiastic areas – NHS Heart of Birmingham – but deals a severe blow to the way the diabetes checks are organised.

The study looked at 34,022 patients entering the NHS Health Check scheme in Birmingham and examined their GP records to see which patients would qualify for a diabetes test according to the ‘diabetes filter’ recommended by the DH.

The filter specifies that patients with a BMI at or above 30 kg/m2 (or 27.5 kg/m2 if of Asian or Chinese ethnicity) or a blood pressure of 140/90 mmHg or higher should be sent for a fasting blood glucose or a HbA1c test.

But researchers discovered that the diabetes filter failed to identify 1,990 (33.3%) of 5,968 patients who had a HbA1c level of 42 mmol/mol or higher recorded within three months of undergoing their Health Check.

They found the diabetes filter had a sensitivity of 67% and a positive predictive value (PPV) of 41% overall, meaning that only two-thirds of those who were actually at risk would have been identified as candidates for blood glucose testing, and less than half would have been found at risk following blood glucose testing.

The authors also noted that Birmingham has a high prevalence of people at high risk of diabetes, so the PPV could be even lower in other, lower prevalence populations.

The team from NHS Birmingham concluded that the whole diabetes arm of the programme should be redesigned to include more modern ways of assessing diabetes risk.

The researchers conclude: ‘Computer-based risk scoring tools for diabetes that have been validated for use in the UK population (as advocated by NICE) may be more effective in risk identification for diabetes.’

Dr Christine D’Acourt, a GP in Oxford and clinical researcher at Oxford University, told Pulse the findings were a cause for concern for the programme running in her area.

She said: ‘A summary report on our NHS Health Checks run in November 2012 showed [in our practice] we had contacted 977 patients, seen 326, and identified an underwhelming five patients at high risk.

‘We had hoped that the programme would at least serve the function of identifying amongst those who attended, any undiagnosed people with diabetes.

‘The recent BMJ Open paper suggests one third of those will be missed, which further erodes the value of the Health Checks.’

She added that the scheme was further exacerbating the inverse care rule in her area and that other outreach methods were needed to identify those at risk.

Dr John Ashcroft, GP in Ilkeston and member of the Derbyshire CHD Committee, said: ‘Health Checks should be calling in those patients off GP registers who are likely to be the most at risk anyway – that is what we are doing locally.’

Click here to read Dr Ashcroft’s full comments

He added that the latest findings suggest that ‘patients should be having HbA1c tested routinely’, but concurred that the test’s specificity and cost-effectiveness really depends on patients’ pretest probability of diabetes – and therefore which patients are selected.

‘I think [the Government] just need to go back and look again at this. I think it’s laudable, but they really should be sticking more closely to what NICE originally said rather than continuing to roll out a system that Gordon Brown thought up one day,’ Dr Ashcroft said.

Summary of main findings  

Patients with HbA1c =42 mmol/molTotalBP/BMI normalPercentage not identified by NHS Health Checks diabetes filter
Ethnicity ‘Asian’3849120731.3%


Pulse Live: 30 April - 1 May, Birmingham

Find out what is new in diabetes screening and treatment from diabetologist Professor Martin Stevens at Pulse Live, Pulse’s new two-day annual conference for GPs, practice managers and primary care managers, will cover the latest developments in telehealth.

Pulse Live offers practical advice on key clinical and practice business topics, as well as an opportunity to debate the future of the profession, and a top range of speakers includes NICE chair designate Professor David Haslam, GPC deputy chair Dr Richard Vautrey and the Rt Hon Stephen Dorrell MP, chair of the House of Commons health committee.

To find out more and book your place, please click here.


Visit Pulse Reference for details on 140 symptoms, including easily searchable symptoms and categories, offering you a free platform to check symptoms and receive potential diagnoses during consultations.