by Colin Kenny, GP, Dromore
It is over 2 years since HbA1c was implemented as our method for the diagnosis of diabetes in the UK. Following World Health Organization recommendations in 2011, an HbA1c of 48 mmol/mol (6.5%) has been the recommended cut point for diagnosis of the condition.
To answer questions on the diagnostic accuracy, cost-effectiveness and guideline utility of HbA1c, Canadian investigators conducted a literature search. The selected studies examined the accuracy of HbA1c compared with fasting plasma glucose or oral glucose tolerance testing (OGTT) for diabetes screening or diagnosis. The majority of papers found that, in comparison with the alternatives, HbA1c testing has a high specificity (true-positive rate) but a lower sensitivity (true-negative rate).
The health technology assessment concluded that absolute values of HbA1c may be more useful as part of an overall diabetes risk assessment, rather than being mutually exclusive from OGTT, owing to the reduced sensitivity.
Interestingly, one of the identified studies showed that a 50g glucose challenge test was the best strategy compared with HbA1c, fasting plasma glucose or OGTT. Given the perceived inaccuracy of this particular assessment when performed in UK general practice, it is unlikely to be re-adopted.
In addition, considerable variation was found in international guideline recommendations for screening for diabetes. Our 2012 NICE guidance on preventing type 2 diabetes recommends that most adults over 40 should be assessed using a validated diabetes risk assessment, and if at high risk be given a fasting plasma glucose or HbA1c test. The guidance goes on to suggest that adults who are in higher-risk ethnic groups or who have conditions that increase the risk of diabetes should be assessed starting at age 25.
To access the full publication, click here.
