The cross-sectional cohort studies in various areas included Yeoncheon county in 1993 with a sample size of 2,520 [9], Jeong-eup district (1997) with a sample size of 1,108 [10], and Mok-dong in Seoul (1998) with a sample size of 774 [11]; the diabetes prevalence rates were 7.2%, 7.1%, and 8.5%, respectively. In a prospective cohort study conducted in Anseong and Ansan counties from 2000 to 2002, 12.6% of residents had been diagnosed with diabetes, including 6.0% established cases of diabetes and 6.6% newly diagnosed cases [12,13]. These cohort studies revealed a rapid increase with respect to the prevalence of diabetes, although the studies were conducted in restricted areas involving participants who might have had widely varied characteristics (Table 1).

The Korean National Health and Nutrition Examination Survey (KNHANES) is a cross-sectional health surveillance system that is randomly sampled throughout South Korea and is conducted by the Korea Institute for Health and Social Affairs. Since its initiation in 1998, surveys for the KNHANES I (1998), KNHANES II (2001), KNHANES III (2005), KNHANES IV (2007 to 2009), and KNHANES V (2010 to 2012) have been executed, and the investigation for the KNHANES VI (2013 to 2015) is still in progress [14]. In this analysis, diabetes was defined as a fasting plasma glucose value ≥126 mg/dL or a previous diagnosis of diabetes. According to the KNHANES, the age-standardized prevalence rates of diabetes among adults aged ≥30 years were 8.6%, 9.1%, 9.6%, 9.6%, 9.8%, and 11.1% in 2001, 2005, 2007, 2009, 2011, and 2013, respectively (Fig. 1A). The prevalence of T2DM was 1.4 times higher in men than in women (Fig. 1A) and increased with age by 2.5% in the group from 30 to 39, 7.3% in people from 40 to 49, 12.6% in those from 50 to 59, 25.2% in the group from 60 to 69, and 27.6% in people 70 years of age and older in 2013 (Fig. 2A). Despite the fact that there are fluctuations in the annual prevalence, the prevalence of diabetes has increased during the last few decades. This rise has been particularly evident in older groups (≥70 years age) and almost doubled from 15.1% in 2001 to 27.6% in 2013.

Furthermore, according to the KNHANES, undiagnosed diabetes was defined as a fasting plasma glucose ≥126 mg/dL without a previous diagnosis of diabetes; the prevalence was 30.9% in men and 22.0% in women out of the total diabetic population in 2013 [14]. The rate of unawareness was highest in people from 30 to 39 years of age and was 65.4% in men and 35.1% in women. Interestingly, according to the 2011 to 2012 KNHANES, it is evident that men with a low income had lower odds for diabetes compared to those with the highest incomes; in contrast, women with a low income had an increased likelihood of developing diabetes compared with women with the highest income [15]. However, according to data from the 2013 KNHANES, the gender-specific economic discrepancy has since disappeared, and diabetes has become more prevalent in individuals in lower income groups in both genders; this finding was similar to that observed in developed Western countries [14].

Recently, the National Health Insurance Service (NHIS), which is a single-payer program for all residents in Korea, signed an agreement with the KDA to provide open access to its databases. As a result, the current epidemic of diabetes was evaluated and reported using the information from the National Health Information Database of the NHIS as a Korean Diabetes Fact Sheet (DFS) in 2015 by the KDA. The data from the NHIS has distinctive characteristics compared to the KNHANES. First, it is not a sampling inventory but instead a complete enumeration survey for all residents in Korea. Second, it is possible to set type limits for diabetes using operational definitions with ICD-10 codes. In practice, patients with type 1 diabetes who had claims under ICD-10 code E10 were excluded from the evaluation of the prevalence of T2DM. According to Korean DFS 2015 [16], about 2.7 million Korean people (8.0%) aged 30 years or older had T2DM in 2013. In this analysis, T2DM was defined based on ICD-10 code (E11-E14) and the concurrent prescription of anti-diabetic medications. The prevalence of T2DM increased with age as follows: 0.9% in the 30 to 39, 3.5% in people 40 to 49, 8.9% in people 50 to 59, 16.6% in the group from 60 to 69, and 21.5% in the 70 to 79 years group; there was a slight decrease in the group ≥80 years to 16.7% in 2013 (Fig. 2B). The overall increase in the annual prevalence of T2DM rose from 5.6% in 2006 to 8.0% in 2013 (Fig. 1B).

When the database of the Health Screening Service was used, which included 10,610,669 participants aged ≥30 years from the national Health Screening Service, the prevalence of diabetes in 2013 was found to be 10.9%; diabetes was defined as a fasting glucose value ≥126 mg/dL or an ICD-10 code (E11-E14) with a prescription of anti-diabetic medications. The prevalence from the database of the Health Screening Service was 2.9% higher than that of the National Health Information Database. This discrepancy suggests that around 3% of the adult population have undiagnosed diabetes according to a fasting glucose ≥126 mg/dL without an ICD-10 code (E11-E14) or the prescription of anti-diabetic medications.