Our results indicate that the prevalence of diabetes is considerably high in the Mudanjiang area of China, especially in the Han population. The prevalences of diabetes in Manchu (8.39%) and Korean Chinese (9.42%) were slightly lower than the national average, while the prevalence of diabetes in Han Chinese in this area (12.10%) was higher than the prevalence in the overall Chinese population (9.7%) [7
]. Prediabetes is an important risk factor for the development of overt diabetes and cardiovascular disease [8
]. The prevalences of prediabetes identified in our study (18.96% in Han, 19.36% in Manchu and 20.47% in Korean Chinese populations) were also much higher than the overall prevalence in China (15.5%). These data have significant public health implications for estimating the burden of diabetes in the Mudanjiang area. The diagnosis of diabetes in the present study was based on both fasting plasma glucose levels and 2-h plasma glucose levels in an OGTT, and the measurement was performed in a qualified central lab. We believe our estimates of diabetes and prediabeteses are accurate.
According to our study, the prevalence of previously undiagnosed diabetes was significantly higher than previously diagnosed diabetes across the three ethnic groups. The proportions of undiagnosed diabetes were 69.94% in the Han population, 88.46% in the Manchu population and 90.48% in the Korean Chinese population, and were much higher than the overall rate in China (60.7%) [7
]. These results raise concerns over the prevention and diagnosis of diabetes in this area, especially in minority groups. Though the prevalence of previously undiagnosed diabetes in the Han population is relatively low, ethnic Han are the major component of the Mudanjiang area population, and the absolute numbers of previously undiagnosed diabetes and prediabetes in the Han population were the largest. Healthcare professionals should work to increase the awareness of diabetes among residents. Active measures should be taken to screen for diabetes.
Diabetes was most frequent among the oldest age groups. Our data showed that the prevalence of diabetes rose with increasing age, and the relationship was steeper in the Han population. However, before being diagnosed with diabetes, a patient might have prediabetes for several years. It is well-known that diabetes can be prevented or delayed with early intervention [10
]. Individuals with prediabetes are at increased risk of developing diabetes [14
]. Therefore prediabetic individuals need attention for diabetes prevention and care. The prevalence of prediabetes among the three ethnic groups differed across age groups. The highest prevalence of prediabetes in the 20-40 year old group was observed in the Korean Chinese population, while in the 40-60 year old group and 60-80 year old group, Han and Manchu Chinese had the highest prevalence of prediabetes among the three ethnic populations, respectively. It is interesting to note that the prevalence of prediabetes is quite stable in Korean Chinese across the different age groups. We speculate that the differences in prediabetic characteristics could be partially attributed to ethnicity, and that this warrants further investigation.
Korean Chinese had higher prevalence of isolated IGT and lower prevalence of isolated IFG than the other two ethnic groups. People with IGT and IFG have different metabolic characteristics. It is reported that individuals with isolated IFG usually have impaired hepatorenal insulin sensitivity, while those with isolated IGT usually have impaired muscle insulin sensitivity [15
]. In Meyer's study, individuals with isolated IGT showed impairments in basal insulin secretion and first-phase insulin release, whereas individuals with isolated IGT showed reduced second-phase insulin release and peripheral insulin resistance [16
]. Further investigations are needed to examine β-cell function and insulin resistance among ethnic populations, which might explain the different profiles of prediabetes among the three ethnic groups.
In our study, the Manchu population had a lower risk of diabetes than the Han population. In the multivariate multinomial logistic regression, the Manchu ethnicity is still an independent protective factor against diabetes. The fact that the demographic and lifestyle factors tested could not explain all of the ethnic differences in diabetes prevalence indicates that other factors such as genetic susceptibility or other endogenous or environmental factors may be responsible for the lowest frequency of diabetes in the Manchu population.
The regression model showed that dietary control and weight monitoring were strongly associated with increased prevalence of diabetes and prediabetes. This might be because people usually begin to pay attention to their diet and body weight after the diagnosis of diabetes or obesity.
This study had several strengths. First, we are the first to present data of the current prevalence of diabetes and prediabetes among the three ethnic groups, Han, Manchu and Korean, in the Mudanjiang area of China. Second, the prevalences of diabetes and prediabetes were different among different age groups and different ethnic populations, and this finding provides fundamental information for the prevention and treatment of diabetes. Finally, that the Manchu ethnicity was a protective factor provides a clue for further investigation of genetic mechanisms.
In the present study, women residents were oversampled, and there was a lower response rate among men than among women. We took these issues into account when we calculated statistical weights.