Our study is one of the first to examine the factors affecting people receiving diabetes education in China using a national representative sample. The results showed great variation in the receipt of diabetes education. Additionally, data from this study revealed that place of residence plays an important part in determining whether one will receive diabetes education and the aspects of diabetes education.

National standard guideline for diabetes education from the American Diabetes Association (ADA) and the China Diabetes Society (CDS) suggest that individualized diabetes education may cover core topics such as nutritional management, physical activities, and preventing, detecting, treating complications [23,24]. Our study indicated that diet and exercise advices were the two most common diabetes education topics among all the five. However, smoking control and foot care education were the lowest two, which are more involving lowering the risk factors that damage blood vessels as well as preventing or delaying complications. Smoking is prevalent in China—the world largest tobacco consumer [25]. However, only 35.12% of the Chinese adults were aware of the health hazards related to smoking. Less than 50% of the current smokers in our study reported receiving quit smoking education, especially among urban residents. Thus, diabetes education regarding avoiding tobacco use is urgently needed. Apart from smoking, our results showed that diabetic patients in China received limited education on foot care, which is consistent with previous study [13]. People with diabetes could develop a series of foot problems including nerve damage, skin changes, and foot ulcer [26]. These complications could be easily prevented by performing regular foot care [26]. Guidelines for standard medical care for diabetes from the ADA and CDS [23,24], recommend that health professionals should provide foot care education to all diabetic patients. As a result, our findings call for future comprehensive diabetes education with a particularly focus on foot care.

Our study showed that residential settings are related to receipt of diabetes education. Migrant population and rural residents compared with urban residents were less likely to receive any diabetes education and less aspects of diabetes education. Diabetic patients living in rural areas and migrant population were less likely to be educated about diet in comparison with their urban counterparts. These results were consistent with our exploratory analysis that identified the association between medical insurance type and receipt of diabetes education, which we found that people with urban employee or government medical insurance were more likely to receive diabetes education about diet, exercise, and foot care. It is possible that these disparities by place residence were partially due to medical insurance type. In comparison with New Rural Cooperative Health Insurance, urban employee and government medical insurance have higher outpatient coverage and help people to access to health services [27,28]. Previous studies have shown that individuals with urban employee and government health insurance were more likely to use outpatient services [27,29]. Diabetes education, as part of chronic disease management, would happen during outpatient visits. As a result, diabetic patients with health insurance that has a higher coverage of outpatient visit would have more outpatient care utilization, and subsequently would be more likely to receive diabetes education.

Besides the differences in health insurance, disparities of health resources may also relate to the receipt of diabetes education. Residents in urban area were more likely to see a doctor in tertiary or secondary hospitals, where diabetes education programs or certified diabetes educators were available. While rural residents and migrant population were more likely to use village clinics, township hospitals, or private clinics that lack of professional doctors and nurses, especially trained diabetic health care providers, which add another barrier of receiving diabetes education [27-29].

Aiming at providing universal health coverage, the Chinese government has launched different public health insurance programs for various populations, with a particular focus on the rural population [28]. Despite the rapid expansion of insurance coverage, with over 96% New Cooperative Health Insurance coverage among rural population, Chinese adults from rural areas still have a relatively high out–of–pocket payment during outpatient visits [28]. Regional disparities of health resources and health professional still existed [28]. Given the great epidemic of non–communicable diseases in China, such as diabetes, there is an urgent need to expand outpatient care coverage for rural population as well as improve quality of health services especially in rural community health centers. The government should focus on increasing reimbursement rates for outpatient care, especially for those from rural areas or migrant populations with chronic diseases. Community capacity building should also be prioritized to promoting non communicable disease prevention and control. In addition to the policy and programs mentioned above, providing flexible reimbursement plans that would cover outpatient care both in rural and urban settings for migrant population may also encourage this segment of the population to use outpatient health promotion services, such as diabetes education classes when needed.

Twenty percent of the diabetics did not receive any diabetes education, which is similar to the rates reported in previous studies [8,13]. Participants with no diabetes education were more likely to be in a migrant population or rural residents, older, or females. These findings were consistent with prior research that age and gender are significant factors associated with not receiving diabetes education. Older adult and females have a higher prevalence of diabetes [27], further diabetes education interventions should be more targeted to reach this group of people.

In our study, no significant association was found between community characteristics (access to health care facility, use of community health facility) and receipt of diabetes education. The lack of strong association could be partially contributed to the fact that most community health centers lack of qualified professional health care providers to provide health education related to prevention and control of non–communicable diseases, such as diabetes [28]. Previous studies provided evidence that the majority of health professionals who worked in primary health care facilities only received 2–3 years basic medical training [30,31]. The shortage of professional trained health care providers could limit the community health centers’ capacity in providing adequate education to tackle the issue of non–communicable disease prevention. Further research is needed to assess whether access to different types of health care facilities is associated with the receipt and quality of diabetes education.

Findings in this study support that contextual factors (age, gender, comorbidities, year of diabetes) were associated with individuals receiving diabetes education, which is consistent with prior research [8,13]. We also found that other contextual factors, an individual’s health behaviors, were associated with the aspects of diabetes education that health professionals provided them. Overweight and obese patients were more likely to receive weight control education. Smokers were more likely to be educated about smoking control. These results indicated that health professionals in China have the tendency to provide diabetes education based on a patient’s health behaviors and health status. However, we still should notice that there was no difference in exercise education among patients with different BMIs. Given the fact that diabetes self–management involves various activities like maintaining a normal body weight, eating healthy diet, and being physical active, failure to provide adequate and comprehensive diabetes education to diabetic patients may increase the risk of adverse health outcomes. With reference to diabetes education program development, our findings suggest that tailored diabetes education covering all aspects of diabetes knowledge are highly needed and would contribute to better diabetes self–management.

This study suffered from several limitations. First, the diagnosis of diabetes is a self–reported measure. It is possible that there is a huge undiagnosed diabetic population, especially in the rural areas. We may have different results if we recruit participants based on medical records or objective measure such as HbA1c. Second, the measure of diabetes education that the CHARLS included is crude so that we are not able to explore the specific education contents. In addition, the guideline from ADA and CDS suggests that diabetes education should include information about disease process, self–monitoring blood glucose, and individualized strategies to address psychological issues [23,24]. Due to limited information collected in the CHARLS data set, we are unable to explore whether these aspects of diabetes education is associated with place of setting. We also did not include health insurance as a contextual factor in our study because of the high colinearity with place of settings. Additionally, the CHARLS data set is cross–sectional. We could only assess the factors that are associated with disparities in receipt of diabetes education due to the nature of the data available. Longitudinal studies are essential to test the causal relationships. Finally, increasing evidence suggests that eye care and dental care are important for diabetes patients; however, the data does not list any of these as a response category. While very little attention has been paid to these care for the patients, practical guidelines and clinical practice needs to be improved to incorporate these as a part of their guideline and routine practice.

Our study found that the receipt of diabetes education was strongly associated with people’s residential locations. Aspects of diabetes education also varied by place of residence in China. Still a large amount of diabetic patients did not receive smoking control and foot care education. Gender, age, years of diabetes, and numbers of comorbidities were significant factors associated with to people receiving different aspects of diabetes education. Individuals health behaviors were also associated the aspects of diabetes education given by health professionals. However, we need to be aware that the results are based on self–reported information on the receipt of diabetes education; thus, the results may be subject to recall bias.

Our study suggests that expanding outpatient care coverage and providing more tailored and comprehensive education are crucial for facilitating diabetes self–management and preventing diabetes complications for middle–aged and older adults in China. In addition, health policies should promote the strengthening of community–based non–communicable diseases health education and health promotion based on community health capacity building. In China, nurses serve as diabetes educators – a critical element in diabetes management. The shortage of nurses and uneven distribution of health care facilities require more adequate training for nurses to be able to educate diabetics self–managing their diseases. In addition, some innovative diabetes education interventions such as mobile health programs are urgently needed as a supplement of health professional diabetes education.