Lack of association between cadmium exposure and risk of diabetes: A longitudinal study from the Chang Gung Research Database.

Abstract

The relationship between chronic cadmium (Cd) exposure and diabetes risk remains uncertain, as previous epidemiological studies have reported inconsistent results. This study aimed to investigate the association between urinary Cd (U-Cd) levels and new-onset diabetes in a large longitudinal Asian cohort. Using data from the Chang Gung Research Database in Taiwan, we included 9576 non-diabetic adults aged 18-65 years who had baseline U-Cd measurements between 2013 and 2020. Participants were stratified into quartiles based on U-Cd concentrations. The primary outcome was new-onset diabetes, assessed using Cox proportional hazards models adjusted for demographic characteristics, body mass index, smoking status, comorbidities, laboratory parameters, and diabetes-related medication use. During a median follow-up of 5.1 years, 585 participants developed diabetes. The crude incidence rates per 1000 person-years across increasing U-Cd quartiles were 9.16, 11.76, 14.97, and 13.10, showing no significant linear trend (P = 0.206). Fully adjusted hazard ratios for quartiles 2, 3, and 4 compared with quartile 1 were 0.85 (95% confidence interval: 0.65-1.11), 1.02 (0.78-1.33), and 0.80 (0.60-1.07), respectively. Restricted cubic spline analyses revealed no significant dose-response relationship across the observed U-Cd range (P = 0.639). These findings were consistent across subgroups defined by sex, age, smoking status, and individuals at high risk for developing diabetes (all P for interaction > 0.05). In conclusion, baseline U-Cd levels were not independently associated with new-onset diabetes in this large Asian cohort. These results suggest that chronic Cd exposure may not be a significant risk factor for diabetes in this population.