van Dam RM, Hu FB, Rosenberg L, Krishnan S, Palmer JR: Dietary calcium and magnesium, major food sources, and risk of type 2 diabetes in U.S. black women. Diabetes Care 29:2238–2243, 2008

The authors state the following with regard to their article:

The results reported in our article were affected by an error in the specification of the age of entry into the study for our Cox proportional hazards regression models. We have repeated our analysis using correct methods, resulting in a weaker association between magnesium intake and diabetes risk. After adjustment for age, BMI, cigarette smoking, physical activity, parental history of diabetes, education level, energy intake, and consumption of alcohol, sugar-sweetened soft drinks, and processed meat and other red meat, the hazard ratio (HR) for type 2 diabetes was 0.87 (95% CI 0.74–1.02; Ptrend = 0.10) for the highest compared with the lowest quintile of magnesium intake. Similar to our results reported in our article, the revised multivariable-adjusted results for calcium (HR 0.92 [95% CI 0.79–1.07] for the top vs. the bottom quintile; Ptrend = 0.11), total dairy (0.90 [0.73–1.11] for ≥2 servings/day vs. <1 serving/week; Ptrend = 0.45), and high-fat dairy (0.98 [0.84–1.15] for ≥1 serving/day vs. <1 serving/week; Ptrend = 0.83) indicated that these were not substantially associated with diabetes risk. The association with low-fat dairy became slightly weaker (0.90 [0.79–1.03] for ≥1 serving/day vs. <1 serving/week; Ptrend = 0.08). Higher whole-grain consumption remained associated with a lower risk of type 2 diabetes (0.82 [0.71–0.94] for ≥1 serving/day vs. <1 serving/week; Ptrend = 0.006).