Between January and July 2022, a multistage stratified sampling method was employed to recruit participants. In the first stage, five cities—Nanning, Guilin, Hechi, Chongzuo, and Yulin—were randomly selected to represent the five geographic regions of the Guangxi Zhuang Autonomous Region (central, eastern, southern, western, and northern). In the second stage, three counties were randomly selected from each city, resulting in 15 counties selected for this study. One hospital (specifically, the internal medicine department) in each county was selected, and 152 patients with diabetes were surveyed per hospital. A total of 2,280 patients were initially enrolled. Inclusion criteria were as follows: (1) diagnosis of diabetes based on the 2020 criteria of the Chinese Diabetes Society (11), (2) age ≥ 18 years, and (3) registered household and current residence within the selected survey area. Exclusion criteria included the following: (1) patients with gestational diabetes; (2) patients with dementia or other psychiatric disorders, as identified through medical records or reported by family members; and (3) patients in the acute phase of illness and unable to cooperate with the investigation. All participating hospital administrators provided institutional consent, and all individual participants provided written informed consent.

Glycemic control was defined according to the Chinese Guidelines for the Prevention and Treatment of Type 2 Diabetes Mellitus (2020 edition) (7) as HbA1c<7.0%. Glycemic control rate was calculated as follows:

Diabetes-related complications included diabetic nephropathy, diabetic foot, lower limb vascular disease, retinopathy, and neuropathy—all confirmed through clinical diagnosis. Smoking was defined as continuous or cumulative smoking for ≥6 months. Categories were defined as follows: regular smoking (≥1 cigarette/day), occasional smoking (>4 times/week but<1 cigarette/day), and former smoker (quit smoking for >6 months). Alcohol consumption was categorized as regular drinking (more than once per week but not daily), occasional drinking (once per week or less), and former drinker (quit for more than 6 months). For analysis, smoking and drinking were classified as “yes” (currently regular use) or “no” (never, former, or occasional use).

The study protocol was approved by the Ethics Committee of the corresponding hospital (approval number: KT-KJT-2021-26). The purpose, voluntary nature, and confidentiality of the study were explained using a questionnaire. All hospital administrators and participants provided informed consent.

Data were collected through in-person questionnaire-based interviews and blood sampling conducted at local hospitals. Investigators provided standardized instructions to guide participants in completing the questionnaires. General information and HbA1c test results from all 15 hospitals were recorded and entered into Excel spreadsheets, which were electronically submitted to the research team. All data were independently verified by two researchers and subsequently imported into SPSS for analysis. Data were stored on password-protected computers to ensure confidentiality.

A database was created using Microsoft Excel, and statistical analyses were performed using SPSS version 25.0. Continuous variables with normal distribution were presented as mean ± standard deviation, and group comparisons were conducted using t-tests. Categorical variables were described as frequencies and compared using the chi-square (χ²) test. Factors associated with HbA1c control were analyzed using multivariate logistic regression. Statistical significance was set at p< 0.05.

A total of 2,280 rural patients with diabetes were surveyed, of whom 2,178 provided valid responses and were included in the final analysis. The participants had a mean age of 63.25 ± 12.71 years and a mean duration of diabetes of 7.96 ± 4.07 years. Among these participants, 55.28% were men, indicating a higher proportion of men than women. Regarding socioeconomic characteristics, 31.43% had completed junior high school, while 37.74% had an annual per capita household income of ≥18,931 RMB. A family history of diabetes was reported by 14.55% of the respondents. For glycemic control, 22.68% of participants had an HbA1c level<7.0%, while 77.32% had an HbA1c level ≥7.0%. Detailed demographic and clinical characteristics are presented in Table 1 .

Among the 2,178 rural patients with diabetes, 494 had HbA1c levels<7.0%, yielding a glycemic control rate of 22.68%. The remaining 1,684 patients (77.32%) had HbA1c levels ≥7.0%. Patients with uncontrolled HbA1c levels were generally older and had a longer duration of diabetes than those with controlled HbA1c levels. Characteristics associated with poorer HbA1c control included being male, having no formal education, being unmarried or widowed, having a lower household income, smoking, alcohol consumption, the presence of diabetes-related complications, the absence of a family history of diabetes, and not receiving oral hypoglycemic agents or insulin therapy (all p< 0.05). Detailed results are presented in Table 2 .

Multivariate logistic regression analysis was performed to identify factors independently associated with HbA1c control, with glycemic control (HbA1c<7.0%) as the dependent variable (0 = controlled, 1 = uncontrolled). Independent variables included age, sex, family history of diabetes, use of oral antidiabetic drugs, insulin therapy, educational level, per capita annual household income, duration of diabetes, and presence of complications. The analysis revealed that age, use of oral antidiabetic drugs, insulin therapy, educational level, income, and duration of diabetes were significantly associated with HbA1c control (p< 0.05) ( Table 3 ). Men had a slightly higher likelihood of achieving glycemic control than did women (OR: 0.800; 95% CI: 0.639–1.002). Each additional year of age was associated with a 2.6% increase in the risk of poor glycemic control (OR: 1.026; 95% CI: 1.017–1.036). Participants with a family history of diabetes were more likely to have poor glycemic control than those without (OR: 1.311; 95% CI: 0.973–1.766), although the difference was not significant. Regarding treatment, both the use of oral antidiabetic drugs (OR: 0.485; 95% CI: 0.377–0.624) and insulin therapy (OR: 0.425; 95% CI: 0.338–0.534) were significantly associated with better glycemic control. Higher educational attainment was associated with improved HbA1c control. Compared with participants who have no formal education, those with primary education (OR: 6.507; 95% CI: 3.076–13.767), junior high school education (OR: 5.557; 95% CI: 2.818–10.955), senior high school education (OR: 2.848; 95% CI: 1.485–5.462), vocational college education (OR: 2.479; 95% CI: 1.285–4.782), and bachelor’s degrees or above (OR: 1.915; 95% CI: 0.943–3.889) had more favorable control rates. Economically, participants with a per capita annual household income ≥18,931 RMB were significantly more likely to achieve HbA1c control than those with lower income levels (OR: 0.662; 95% CI: 0.528–0.830). Additionally, each 1-year increase in diabetes duration was associated with a 13.7% increase in the risk of poor glycemic control (OR: 1.137; 95% CI: 1.104–1.171). No significant association was found between the presence of diabetes-related complications and HbA1c control (OR: 0.945; 95% CI: 0.739–1.209; p = 0.652).

This study had a few limitations. First, it was a multicenter cross-sectional survey conducted in the internal medicine departments of 15 county-level hospitals in the Guangxi Zhuang Autonomous Region. All participants were patients who actively sought medical care, excluding undiagnosed or untreated individuals with diabetes in rural areas. This introduces a selection bias, as those who visit hospitals may exhibit greater health literacy and better treatment adherence. Therefore, the reported HbA1c control rate may have been overestimated and should not be directly generalized to the broader rural diabetic population in Guangxi.

Second, the sampling process involved randomly selecting one prefecture-level city from each of the five geographical regions of Guangxi (central, eastern, southern, western, and northern), followed by the random selection of three counties within each city. Although this approach aimed to account for geographic and healthcare diversity, it did not incorporate stratified weighting based on rural population proportions. Consequently, areas with larger rural populations were not given proportionally greater sampling weights. Thus, the sample may not fully represent all rural patients with diabetes in the region, and the findings primarily reflect those who actively seek medical care.

Third, because of practical constraints, the types of HbA1c testing equipment used varied across participating hospitals. Although all sites utilized high-performance liquid chromatography, and testing was conducted by trained laboratory professionals, minor measurement variability between instruments may persist. To minimize potential bias, we adopted an HbA1c threshold of<7.0%, as recommended by the Chinese Guidelines for the Prevention and Treatment of Type 2 Diabetes Mellitus (2020 edition). Nevertheless, caution is advised when interpreting the HbA1c control rate data.

The HbA1c control rate among rural patients with diabetes in the Guangxi Zhuang Autonomous Region remains suboptimal. Targeted attention should be directed toward individuals who are older, have a longer duration of diabetes, are receiving oral hypoglycemic or insulin therapy, have lower educational levels, or belong to low-income households. Tailored lifestyle interventions, along with enhanced training of community healthcare providers, are essential. Improving patients’ health literacy and reinforcing their awareness of and responsibility for diabetes self-management may enhance adherence to follow-up care and lead to better outcomes in community-based diabetes control programs.