Abstract
BACKGROUND: The relationship between oxidative balance score (OBS) and diabetes remains poorly understood and may be gender-specific. We conducted a cross-sectional study to investigate the complex association between OBS and diabetes among US adults. METHODS: Overall, 5,233 participants were included in this cross-sectional study. The exposure variable was OBS, composed of scores for 20 dietary and lifestyle factors. Multivariable logistic regression, subgroup analysis, and restricted cubic spline (RCS) regression were applied to examine the relationship between OBS and diabetes. RESULTS: Compared to the lowest OBS quartile group (Q1), the multivariable-adjusted odds ratio (OR) (95% confidence interval (CI) for the highest OBS quartile group (Q4) was 0.602 (0.372-0.974) (p for trend = 0.007), and for the highest lifestyle, the OBS quartile group was 0.386 (0.223-0.667) (p for trend < 0.001). Moreover, gender effects were found between OBS and diabetes (p for interaction = 0.044). RCS showed an inverted-U relationship between OBS and diabetes in women (p for non-linear = 6e-04) and a linear relationship between OBS and diabetes in men. CONCLUSIONS: In summary, high OBS was negatively associated with diabetes risk in a gender-dependent manner.