Total Iron Bioavailability from the US Diet Is Lower Than the Current Estimate

Background: Total (heme and nonheme) iron bioavailability from the US diet has been estimated to be 18% based on a single human absorption study. New data, however, suggest that it may be time to revisit this estimate. Objective: We estimated total iron bioavailability from the US diet with the use of our recently reported algorithm that estimates nonheme iron absorption and a conservative value for heme iron absorption. Methods: We used dietary intake and biomarker information from the NHANES 2001-2002, My Pyramid Equivalents Database, and Food and Nutrient Database for Dietary Studies. The survey package in R software was used to estimate means and CIs, taking into account the strata, primary sampling units, and appropriate survey weight. We implemented 2 different approaches to estimate total iron absorption. In the first approach, we included all survey participants but adjusted the geometric mean of nonheme iron absorption to 15 mu g ferritin/L serum to mimic values of individuals with no iron stores; in the second approach, absorption was estimated for only nonanemic subjects with no iron stores. A total sample size of 6631 was used based on availability of dietary and iron status biomarker data and C-reactive protein concentration <= 6 mg/L Results: The geometric mean (95% CI) of unadjusted nonheme iron absorption for all subjects was 3.7% (3.6%, 3.8%), higher in female subjects [5.6% (5.4%, 5.7%)] than male subjects [2.6% (2.5%, 2.7%)] (P < 0.0001). Nonheme iron absorption was lower in non-Hispanic whites [3.5% (3.4%, 3.6%)] than Mexican Americans [4.5% (4.2%, 4.8%)] and non-Hispanic blacks [4.4% (4.1%, 4.7%)]. Estimated total iron absorption was 15.5% or 15.1%, depending on which approach was used to carry out the calculations. Conclusion: This study provides useful data for evaluating the current value of iron bioavailability from the US diet.