Background Diabetes is the tenth leading cause of death worldwide, posing a major threat to global public health. In recent years, occupational factors—particularly noise exposure—have been increasingly recognized as potential contributors to diabetes. However, current evidence linking occupational noise exposure to diabetes is limited and the findings are inconsistent.

Objective To investigate the impact of occupational noise exposure on fasting blood glucose levels, impaired fasting glucose (IFG), and the risk of diabetes among workers, to identify high-risk populations, providing scientific insights for diabetes prevention and control in occupational settings.

Methods This study was conducted in 2023 among workers in the automotive manufacturing industry in Guangzhou. Individual occupational noise exposure was evaluated using cumulative noise exposure (CNE). Fasting blood glucose was measured during physical examinations, and relevant baseline information was collected through questionnaire surveys and physical examinations. Linear regression, logistic regression, restricted cubic spline (RCS) were used to analyze the association between occupational noise and fasting blood glucose level, IFG and diabetes. Subgroup analysis and interaction test were performed to identify high-risk populations.

Results A total of 3427 participants from two factories in Guangzhou were included finally. The average age of the participants was 28.0 (8.0) years, with a median CNE level of 86 dB·years. There were 190 cases of IFG and 36 cases of diabetes. Using continuous variables modeling, the results showed associations between CNE and blood glucose levels (b=0.006, 95%CI: 0.002, 0.011) as well as diabetes (OR=1.145, 95%CI: 1.039, 1.270). High noise exposure significantly elevated the risk of diabetes in the population (OR=3.908, 95%CI: 1.255, 17.341) in the models using categorical variables. The RCS analysis showed a linear dose-response relationship between occupational noise and diabetes (P for overall=0.048, P for non-linear=0.567). Age was a significant modifier in the association between occupational noise and blood glucose levels (P for interaction<0.001), with the association observed to be more pronounced among workers aged >28 years compared to those aged ≤28 years (b=0.017, 95%CI: 0.009, 0.025). However, no significant interaction was found between IFG and diabetes.

Conclusion Occupational noise exposure may increase the risk of diabetes, with a linear dose-response relationship. Workers aged over 28 years are at high risk. Hence, strengthening workplace noise control measures and raising awareness among workers are essential steps to mitigate the adverse health effects of occupational noise exposure.