Diabetes risk scores in 2011

Abstract

Diabetes risk scores can be used as pre-screening tools to detect those likely to have diabetes. Scores usually include clinical characteristics such as age, sex, family history of diabetes and hypertension. However, it is disputed whether screening for diabetes is cost-effective. The recently reported Anglo-Danish-Dutch Study of Intensive Treatment in People with Screen Detected Diabetes in Primary Care (ADDITION) study, in which diabetes was diagnosed following screening by a risk score, did not show that intensive treatment in such individuals was different from routine care in terms of cardiovascular outcomes. Risk scores are also used to identify those at risk of diabetes in the future, and at-risk individuals may then be encouraged to participate in diabetes prevention programmes. Risk scores from routine biology, in particular fasting glucose, have also been developed to improve prediction over clinical risk factors. Now more sophisticated approaches are being used to predict diabetes - multiple biomarkers, genetics, proteomics, lipidomics and metabolomics - with the idea that if individuals are identified a long time in advance of the onset of the disease, prevention can start much earlier when it may be more successful. Diabetes risk scores follow on from a long history of cardiovascular risk scores. Scores should be given with an uncertainly or prediction interval within which the score lies with 95% confidence. © TOUCH BRIEFINGS 2011.