BACKGROUND: Studies document that magnesium is inversely associated with the risk of diabetes, which is a risk factor of pancreatic cancer. However, studies on the direct association of magnesium with pancreatic cancer are few and findings are inconclusive. In this study, we aimed to investigate the longitudinal association between magnesium intake and pancreatic cancer incidence in a large prospective cohort study. METHOD: A cohort of 66,806 men and women aged 50-76 years at baseline who participated in the VITamins And Lifestyle (VITAL) study was followed from 2000 to 2008. Multivariable-adjusted Cox regression models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) of pancreatic cancer incidence by magnesium intake categories. RESULT: During an average of 6.8-year follow-up, 151 participants developed pancreatic cancer. Compared with those who met the recommended dietary allowance (RDA) for magnesium intake, the multivariable-adjusted HRs (95% CIs) for pancreatic cancer were 1.42 (0.91, 2.21) for those with magnesium intake in the range of 75-99% RDA and 1.76 (1.04, 2.96) for those with magnesium intake <75% RDA. Every 100 mg per day decrement in magnesium intake was associated with a 24% increase in the incidence of pancreatic cancer (HR: 1.24; 95% CI: 1.02, 1.50; P(trend)=0.03). The observed inverse associations appeared not to be appreciably modified by age, gender, body mass index, and non-steroidal anti-inflammatory drug use but appeared to be limited to those taking magnesium supplementation (from multivitamins or individual supplement). CONCLUSIONS: Findings from this prospective cohort study indicate that magnesium intake may be beneficial in terms of primary prevention of pancreatic cancer.