Contributions of total daily intake of folic acid to serum folate concentrations.

To the Editor: Recent literature has associated potential adverse effects with daily use of folic acid supplements (1000 μg), high concentrations of serum folate (26.5 ng/mL or greater [to convert to nmol/L, multiply by 2.266]), or high folic acid intake, but researchers have not adequately considered the source of folic acid. We examined the relative contribution of different sources of folic acid to serum folate concentrations among US adults. Methods. Data were obtained from a single 24-hour dietary recall from the National Health and Nutrition Examination Survey (NHANES) 2001 through 2004 for 8655 nonpregnant participants aged 19 years and older who also had serum folate concentrations measured. The 24-hour dietary recall in NHANES underestimates energy intake by 11% among adults. Daily folic acid intake from individual fortified foods was calculated. Average daily intake of folic acid from supplements over the prior 30 days was calculated. The 3 main sources of dietary folic acid are enriched cerealgrain products (ECGPs) fortified at 140 μg of folic acid per 100 grams of flour, ready-to-eat cereals (RTEs) such as breakfast cereals with up to 400 μg of folic acid per serving, and supplements containing folic acid. Serum folate concentration was log-transformed because the distribution was not Gaussian. For each serum folate quintile, the adjusted geometric means for serum folate concentration and adjusted mean daily intake of folic acid from ECGPs, RTEs, and supplements containing folic acid were estimated using multiple regression models. Daily intake of folic acid from ECGPs only was estimated by subtracting the intake of folic acid from RTEs from the intake of folic acid from all foods. The proportion of participants who reported using a supplement containing folic acid in the past month was estimated using a logistic regression model. Differences across serum folate quintiles for each intake source and for proportion of supplement users were tested using Satterthwaite adjusted F statistic. All analyses were weighted, taking into account the complex sampling design of NHANES, using SUDAAN version 9.0 (RTI, Research Triangle Park, North Carolina), and adjusted for age, sex, and race/ethnicity. In NHANES, race/ethnicity is designated by the participant through a series of multiplechoice questions and an “other” write-in response. Statistical hypotheses were tested at a 2-tailed =.05 level of significance. Results. Participants in the highest serum folate quintile were more likely to be older, female, and non-Hispanic white (P .001 for each comparison) (TABLE). Across the serum folate quintiles, the daily intake of folic acid from ECGPs was about 140 μg per day and showed a statistically significant but small variation (P=.01) (FIGURE). However, from the first to the fifth quintile, daily intake from RTEs varied from 14 to 107 μg per day (P .001) and daily intake from supplements varied from 42 to 392 μg per day (P .001). The adjusted proportion of ECGP-only users varied from 82.1% to 13.9% (P .001) and supplement users varied from 10.9% to 74.6% (P .001), respectively (Table).