BACKGROUND AND AIM: While selenium (Se) is an essential nutrient, some human studies showed an association between increased Se exposure and diabetes. However, studies evaluating the role of Se in insulin resistance and beta cell function are limited. We evaluated the cross-sectional association of Se exposure with homeostasis model assessment of insulin resistance (HOMA-IR) and β-cell function (HOMA-B), two established predictors of diabetes susceptibility. Because this relationship may change with age, we conducted the analyses among middle age and older adults separately. METHODS: We included 1161 participants from the Aragon Workers’ Health Study (AWHS) and 915 from Seniors-ENRICA-2 (SEN-2) study. All participants were diabetes-free. Se exposure was measured in blood for AWHS and in serum for SEN-2. HOMA-IR and HOMA-B were calculated using fasting glucose and insulin levels. RESULTS:Median age in the AWHS and SEN-2 were 52 and 70 years, respectively. Median levels of blood and serum Se in AWHS and SEN-2 were 214 and 115.5 μg/L, respectively. In AWHS, the geometric mean ratio (GMR, 95% confidence interval) comparing the 90th and 10th percentiles of Se distribution was 1.07 (0.99, 1.15) for HOMA-IR and 1.14 (1.06, 1.24) for HOMA-B. In SEN-2, the corresponding GMR (95% CI) were 1.10 (1.00, 1.21) for HOMA-IR and 0.93 (0.84, 1.03) for HOMA-B. In a subsample of 611 AWHS participants with available follow-up the shape of the dose-response of selenium with prospectively measured HOMA-IR and HOMA-B was consistent compared to the cross-sectional dose-response. CONCLUSIONS:The positive association of Se with HOMA-IR in both studies suggests that high levels of Se exposure are related with increased insulin resistance. Alternatively, the positive association of Se with HOMA-B in middle aged adults from AWHS but not in older adults from SEN-2, might suggest that increased insulin resistance induces compensatory increased β-cell function in younger ages, being this compensatory capacity decreased with aging. KEYWORDS: Epidemiology, obesity and metabolic disorders, exposures