Aims: To estimate the prevalence of diabetes mellitus with three diagnostic criteria (WHO-1985 and 1999 and ADA-1997), evaluate their concordance and analyse the sensitivity and specificity of the different screening strategies for diabetes.
Methods: A cross-sectional population study with two-step sampling. One thousand and 34 people were selected randomly. A 75-g oral glucose tolerance test (OGTT) was performed and venous blood samples were obtained fasting and at 2 h.
Results: The prevalence of known Type 2 diabetes mellitus (DM-2) is 4%[95% confidence interval (CI) 2.8, 5.1]. By WHO-1985 criteria the prevalence of unknown DM-2 is 5.9% (4.5, 7.4); by ADA-1997 criteria 3.5% (2.5, 4.6) and by WHO-1999 criteria 7.3% (5.8, 8.8). Diagnostic overlap and statistical concordance (coefficient K) are WHO-1985/ADA-1997 29.3%, K=0.42; WHO-1985/WHO-1999 80%, K=0.88; ADA-1997/WHO-1999 48%, K=0.63. If only fasting glucose was used (following ADA-1997), 36.3% of those with diabetes (2-h glucose > or =11.1 mmol/l) would be diagnosed. If OGTT was performed (i) in those with a fasting glucose between 6.1 mmol/l and 6.9 mmol/l (9.8% of the population) we would diagnose 66.6%, and (ii) in all those between 5.7 mmol/l and 6.9 mmol/l (18.9% of the population) 81.8% would be diagnosed.
Conclusions: The ADA criteria decrease the prevalence of DM in the adult population of Asturias by 2.4% and concordance with the classical criteria (WHO-1985) was only 29.3%. Using fasting glucose only (ADA-1997) diagnoses 36.3% of those with diabetes. The recent recommendations of the WHO-1999 increases this to 66.6%. To improve the diagnostic strategy for diabetes and detect up to 81.8% of patients, we propose the use of OGTT for all those with a fasting glucose between 5.7 mmol/l and 6.9 mmol/l.