To the best of our knowledge, this present study is the first to report regional differences, especially differences between two provinces, of diabetes prevalence and awareness in China. Results showed both diabetes prevalence and awareness were higher in coastal provinces and lower in interior provinces in China.
Not surprisingly, longitudinally, diabetes prevalence in our study was higher in Fujian Province and Shaanxi Province than that reported in previous literature . Furthermore, like Germany , China also showed difference in diabetes prevalence between large regions. We therefore analyzed risk factors of diabetes prevalence in the two provinces. We found the same factors included age, central obesity [18–20], family history and hypertension [1, 21] in both provinces. However, we could not discover that education level and yearly family income were independently associated with diabetes prevalence as previously reported [22, 23]. Furthermore, we revealed lifestyle risk factors contributed differently to diabetes prevalence in the two provinces. Physical activity was associated with the prevalence in Shaanxi, a finding in agreement with previous studies [24, 25]. But in Fujian, cigarette smoking, not physical activity, was associated with prevalence, which was similar to Morimoto’s study . In addition, considering all affecting factors that we knew, regional significance still remained for diabetes prevalence. Results indicated that these differences could not be accounted for by differences such as population age, sex, ethnics, education level, suggesting that other unknown factors might be at stake. This may partly be explained by air pollution , vitamin D level , or local food economy  between two provinces.
As for awareness,we found the rates of diabetes awareness were higher in Fujian Province and Shaanxi Province than those in the national representative data from 2001 (23.66%)  and those in the non-national representative data from 1998 (33.3%) , similar to those in other developing countries  but lower than those in developed countries [21, 31]. Comparisons between both provinces indicated that the coastal province had relatively higher diabetes awareness than the interior province. Analysis of the factors impacting awareness in the two provinces found that family history of diabetes was the only independent risk factor of diabetes awareness and the regional difference was not significant after adjusting all possible risk factors we knew. Unlike previous reports, for example, Kaiser et al. reported increasing age was positively associated with awareness of type 2 diabetes . Sims et al. reported that socioeconomic status was highly associated with awareness . However, Harwell et al. also found that family history was the factor most significantly associated with the perceived risk of developing diabetes . Our conclusions may be due to China’s specific national situation. In China, the tie among family members is closer. Family members actively participate in the care of subjects with diabetes by accompanying them to health care, by contributing financially for drugs and examination. These surely enhance their level of diabetes awareness. Therefore, individuals with a positive family history of a disease may develop a personal sense of vulnerability, which in turn may increase their awareness.
Our study was a well-designed large representative population-based investigation and OGTT was used, which made the findings more convincing. However, the design fault as a cross-sectional study in disease causality should be considered, especially referring to the risk factors related to diabetes prevalence and awareness. Besides, the principal limitation of the present study was the potential selection bias, because women and urban residents were oversampled and there was a lower response rate in men than in women as described in Yang’s report . Another limitation was that the data of occupation and personal income were lacked and the sample seemed insufficient to provide an urban–rural specific multivariate model. Finally, the limitation should be considered whether the findings about regional differences of diabetes prevalence in our study could be generalized to the whole country needs further investigations, as differences among provinces in China are extremely complex and there are too many risk factors.