Prevalence of Prediabetes and Undiagnosed Diabetes in Canada (2007-2011) According to Fasting Plasma Glucose and HbA1c Screening Criteria

OBJECTIVETo provide the first population-based estimates of prediabetes and undiagnosed type 2 diabetes prevalence in Canada.RESEARCH DESIGN AND METHODSWe combined two fasting subsamples of the Canadian Health Measures Survey, which were restricted to nonpregnant adults 20 years of age (N = 3,494). Undiagnosed diabetes was defined as not having self-reported type 2 diabetes but having blood glucose measures that met Canadian guidelines (i.e., fasting plasma glucose [FPG] level of 7.0 mmol/L or hemoglobin A(1c) [HbA(1c)] level of 6.5% [48 mmol/mol]). Prediabetes was defined as an FPG level of 6.1 and <7.0 mmol/L or an HbA(1c) level of 6.0% and <6.5% (42 and <48 mmol/mol). All estimates were weighted using survey sampling weights. CIs were calculated with the bootstrap method.RESULTSAccording to FPG levels, the prevalence of undiagnosed type 2 diabetes in Canadian adults was 1.13% (95% CI 0.79, 1.62), contributing to approximate to 20% of total type 2 diabetes prevalence (5.62 [95% CI 4.52, 6.95]). Compared with FPG levels, the undiagnosed prevalence was greater using HbA(1c) level as a criterion (3.09% [95% CI 1.97, 4.81]), approximate to 41% of the total number of cases of diabetes (7.55 [95% CI 5.98, 9.49]). The HbA(1c)-only criterion resulted in a threefold increase in prediabetes prevalence overall and a sixfold increase among females (FPG 2.22%, HbA(1c) 13.31%). Screening based on FPG only identified older undiagnosed case patients, with a mean age of 58.7 years (95% CI 59.9, 63.4). Similarly, using HbA(1c) identified younger individuals with prediabetes, with reduced BMI and waist circumference compared with FPG levels.CONCLUSIONSIn this first study of a nationally representative sample with biospecimen measures, we found that the prevalence of undiagnosed type 2 diabetes and prediabetes was significantly higher using HbA(1c) levels compared with FPG levels. Further evaluation is needed to fully assess the impact of using the HbA(1c) criterion.