Folic Acid Intake from Fortification in United States Exceeds Predictions

Abstract

In 1996, the U.S. Food and Drug Administration issued a regulation requiring that all enriched cereal-grain products be fortified with folic acid by January 1998. An average increase in folic acid intake of 100 μg/d was projected as a result of this fortification. The objective of the present study was to estimate the effect of this fortification on the intake of folic acid and total folate, and on the prevalence of individuals with inadequate folate intake and with high folic acid intake. We used data on food and nutrient intake from 1480 individuals who participated in the 5th and 6th examinations of the Framingham Offspring Cohort Study. Fortification was instituted during the 6th examination so that 931 participants were examined before its implementation (nonexposed) and 549 after implementation (exposed). Published data on total folate in enriched cereal-grain products were used to correct folate content in these foods to reflect fortification. Among nonsupplement users, folic acid intake increased by a mean of 190 [95% confidence interval (CI): 176, 204] μg/d (P < 0.001) and total folate intake increased by a mean of 323 (95% CI: 296–350) μg dietary folate equivalents (DFE)/d (P < 0.001) in the exposed participants. Similar increases were seen among supplement users exposed to fortification. The prevalence of exposed individuals with total folate intake below the estimated average requirement (320 μg DFE/d) decreased from 48.6% (95% CI: 44.2–53.1%) before fortification to 7.0% (95% CI: 3.1–10.9%) after fortification in individuals who did not use folic acid supplements. This prevalence was ∼1% or less for users of supplements both before and after fortification. Prevalence of individuals with folic acid intake above the upper tolerable intake level (1000 μg folic acid/d) increased only among supplement users exposed to fortification (from 1.3 to 11.3%, P < 0.001). No changes in folic acid intake were observed over time in the nonexposed participants. By these estimations, folic acid fortification resulted in a mean increase in folic acid intake that was approximately twice as large as previously projected.

folate, folic acid, fortification, folate intake, humans

Observational studies, nonrandomized intervention studies and randomized controlled trials have demonstrated that folic acid taken during the periconceptional period decreases the risk of pregnancies affected by neural tube defects (NTD)⁴ (1–7). In September 1992, the U.S. Public Health Service issued a recommendation that all women of childbearing age capable of becoming pregnant consume 400 μg of folic acid/d to decrease the risk of having an NTD-affected pregnancy (8). In the United States, the level of awareness of the above-mentioned recommendation among women of childbearing age was low, and the proportion of these women who reported taking folic acid-containing supplements was also low (9,10). Fortification of the food supply with folic acid was considered to be the most effective approach to help women of childbearing age increase their periconceptional folic acid intake (11). Thus, in March 1996, the U.S. Food and Drug Administration (FDA) issued a final rule, effective January 1998, that required all enriched cereal-grain products (flour, rice, breads, rolls and buns, pasta, corn grits, corn meal, farina, macaroni and noodle products) to be fortified with folic acid in addition to the iron, thiamin, riboflavin and niacin already added to these enriched cereal-grain products (11).

The amount of folic acid added to different products ranges from 95 to 309 μg/100 g of product. This range of fortification was selected on the basis of a target level of 140 μg folic acid/100 g of the cereal-grain product. The projected average increase of folic acid intake in the general population attributable to fortification of enriched cereal-grain products was estimated to be ∼100 μg/d (12). This projection assumed that each enriched cereal-grain product contained the amount of folic acid required by the applicable regulation. However, actual measurements of total folate content in enriched cereal-grain products have shown that a considerable proportion of these products contained total folate levels that were higher than the amounts required by regulations (13).

Enriched cereal-grain products are fortified with folic acid, the synthetic form of folate, which is more bioavailable than most naturally occurring folates (14). The greater bioavailability of folic acid resulted in the introduction of the concept of “dietary folate equivalent” (DFE) into the dietary reference intakes (DRI) as a way to express a mixed intake of folic acid and natural folate. Accordingly, 1 μg of natural folate equals 1 μg of DFE, and 1 μg of folic acid equals 1.7 μg of DFE (14,15). Thus, it was deemed necessary to distinguish between these two forms of folate to assess the adequacy of folate intake in an individual or a population group. It is also very important to be able to quantify total folic acid intake because the tolerable upper intake levels (UL) are expressed for folic acid alone (14).

The purpose of the current study was to estimate the effect of the FDA-mandated folic acid fortification on folic acid and total folate intake, and on the prevalence of individuals with inadequate folate intake and with high folic acid intake. To accomplish this, we used data on food and nutrient intake from the Framingham Offspring Study and published data on total folate measured in enriched cereal-grain products after implementation of fortification and in other previously fortified products including breakfast cereals (13) to correct the folate values for these foods.

SUBJECTS AND METHODS

Study population.

The Framingham Heart Study (FHS) is an epidemiologic study of heart disease established in Framingham, MA between 1948 and 1950. This cohort originally included 5209 individuals and among them, 1644 husband and wife pairs. Their offspring and the offspring's spouses were recruited and invited to participate in the Framingham Offspring Study. Offspring who had only one parent participating in the FHS were also invited to participate with their spouses if that parent had either abnormal lipoprotein patterns or coronary heart disease at the 1970 biennial examination of the FHS. The first examination of the offspring cohort was in 1971, and they have typically been examined every 3–4 y. The ethnicity of the participants is almost exclusively non-Hispanic white (16).

Data from the 5th and 6th examinations of the Framingham Offspring Cohort were used for this study. The 5th examination took place between January 1991 and December 1994 before implementation of folic acid fortification. The 6th examination started before fortification was implemented (January 1995) and finished after fortification was in place (August 1998). There were 3386 individuals who participated in both examinations. Of these participants, 2818 had valid dietary data at both examinations. Participants were excluded if they either started or stopped using vitamin supplements between the exams (n = 661) or were missing data (n = 15). Finally, participants were selected, and exposure to fortification determined, on the basis of the 6th examination date. From the remaining participants, we selected 931 individuals who attended their 6th examination before enriched grain products fortified with folic acid were introduced into the food supply in the Framingham area and 549 individuals who attended their 6th examination after most enriched grain products were fortified with folic acid. Participants seen during the phase-in period for fortification (n = 662) were excluded.

Details of the selection of both groups are explained elsewhere (17). Briefly, the folic acid fortification final rule allowed food manufacturers almost 2 y (March 1996–January 1998) in which to change the formulations of their products and to provide new labels reflecting the changes. In Southern New England, there were very few products fortified before September 1996 but most products were fortified by July 1997 (17). Therefore, participants of the Framingham Offspring Study whose 6th examination occurred between January 1995 and September 1996 were considered as the group that was examined before implementation of folic acid fortification or group not exposed to fortification and will be referred to as the nonexposed group. Those who attended the 6th examination between September 1997 and August 1998 were considered as the group that was examined after implementation of folic acid fortification or group exposed to fortification and will be referred to as the exposed group. Individuals who were examined in between those dates (October 1996–August 1997) were not considered in the analysis. Thus, we had available dietary data for the exposed and nonexposed groups at two points in time, baseline (5th examination) when fortification was not in place and follow-up (6th examination) when fortification was not in place at the time the nonexposed group was examined but was in place at the time the exposed group was examined. The availability of these data allowed us to compare both groups at baseline to assess the comparability between the groups, before the exposed group was exposed to fortification. It also allowed us to check for differences between groups at follow-up, after the exposed group was exposed to fortification, and to test for differences over a similar period of time within each group. In this way, we were able to assess changes in folate intake in the exposed group as a result of fortification, adjusting for any changes in the nonexposed group unrelated to fortification. This study was approved by the Human Investigations Review Committee at New England Medical Center and by the Institutional Review Board for Human Research at Boston University Medical Center.

Dietary data.

Individuals who participated in the 5th and 6th examinations of the Framingham Offspring Cohort received and completed a 126-item semiquantitative food-frequency questionnaire (FFQ) developed by Willett et al. (18) that allowed for the estimation of usual nutrient intakes during the year before the examination. This FFQ was found to be reproducible and to provide a valid estimate of folate intake over a 1-y period (18,19). The FFQ, which was designed to be self-administered, was completed and returned by each subject at the time of the clinic examination. Subjects were excluded from the analysis if their FFQ had >12 items left blank or if their total energy intake calculated by the questionnaire was <2.51 MJ/d for men and women, >16.74 MJ/d for women or >17.57 MJ/d for men.

The FFQ contained questions about supplement consumption, including frequency of consumption and supplement brand and type. With this information, we were able to identify the contribution of folic acid from supplements and the contribution of total folate from foods to total folate intake. From the total folate intake from foods, we were interested in differentiating between natural folate present in foods and folic acid added to foods. At the time the FFQ were analyzed, the food composition database used to calculate nutrient intakes had not been modified to reflect fortification of enriched cereal-grain products. For each food, folate content present in the food composition database will be referred as “original folate content.” To recalculate folate intake in the exposed and nonexposed groups, we used data from a recently published study in which folate content was measured in >150 enriched cereal-grain products and other products fortified with folic acid including breakfast cereals (13). Although these products were selected on the basis of market data and were the top-ranked products covered by the fortification regulations, they represented a small sample of foods collected in one metropolitan area at one time point. Folate values derived from this study will be referred as “measured folate content.”

To recalculate folate intake in the exposed group after implementation of fortification (follow-up), we identified the foods from the FFQ that were enriched with folic acid as a result of the FDA mandatory fortification. We had data on measured folate content in the following enriched cereal-grain products: bread, corn grits, rice, pasta and corn meal [see (13), Table 12]. For other enriched cereal-grain products, we did not have direct measurements of total folate content. For commonly consumed products, including muffins, pancakes, crackers, pizza, cookies, brownies, doughnuts, cakes, sweet rolls and pie, the measured folate content was estimated on the basis of the amount of flour present in each food product and the measured folate content in fortified flour (20). For infrequently consumed products such as cream of wheat, tortilla, corn bread, stuffing, breadsticks, grits and croutons, we used total folate content from the USDA food composition database (20), which was updated after fortification on the basis of the average theoretical folic acid value required by regulation. Because these foods represent <1% of the total folate intake, the use of theoretical values should have minimal effect on our calculations. For all of these enriched cereal-grain products, original folate content values used to analyze the FFQ were replaced by the measured folate content values (or the content from the USDA food composition database); the original folate content was considered to be natural folate and the difference between measured folate content of the fortified food and the original folate content was considered to be folic acid.

We also had data available on folate content for some foods that were fortified with folic acid before the fortification of enriched cereal-grain products was implemented, i.e., cereal bars, toaster pastries and different classes of fortified breakfast cereals [see (13), Tables 9 and 10]. Consequently, the following modifications were carried out to recalculate folate intake in all individuals (nonexposed and exposed groups) at baseline and follow-up. For cereal bars and toaster pastries, the original folate content data were replaced by the measured folate content. We identified from the FFQ the different classes and brands of breakfast cereals and classified them as fortified (>50 μg total folate/serving size) or not fortified (≤50 μg total folate/serving size) with folic acid on the basis of their original folate content. For fortified breakfast cereals, we replaced the original folate content data by the measured folate content that best matched each particular cereal. An average amount of 31 μg of natural folate/100 g cereal was calculated using data on folate content in nonfortified breakfast cereals. The difference between measured folate content of each fortified breakfast cereal, breakfast bar and toaster pastry and this average amount assigned as natural folate was considered to be folic acid. For nonfortified breakfast cereals, folate data were not modified and 100% of the folate was considered to be natural folate. For the breakfast drinks, Ensure and Ultra Slim Fast, the original folate values were considered to be 100% folic acid. For all of the remaining foods, folate content was not modified and considered to be 100% natural folate.

After incorporating the above-mentioned modifications, we created new variables representing estimations of total folate intake as μg of DFE, μg of natural folate intake and μg of folic acid intake (including both that added to foods and that taken as vitamin supplements).

Laboratory methods.

As part of the 5th and 6th offspring cohort examinations, blood samples were obtained from subjects who had fasted for at least 10 h. Plasma vitamin B-12 (cobalamin) was measured by a radioassay (Biorad Quantaphase II, BioRad, Hercules, CA). The CV for this assay was 7%.

Statistical methods.

We performed our analysis separately for individuals who did and did not use B vitamin supplements, with supplement use defined as taking a multivitamin containing folic acid at least once per week. Total folate, natural folate and folic acid values were positively skewed; thus, square-root transformations were used to normalize these data, and means based on square-root transformations are reported. To estimate the effect of folic acid fortification on folate intake in this cohort, we compared total folate, natural folate and folic acid means between the exposed group and nonexposed groups at baseline and follow-up to assess differences between the groups, and we measured the changes in total folate, natural folate and folic acid within the exposed and nonexposed groups between the two time points (baseline and follow-up). All of these comparisons were carried out using SAS PROC MIXED program with repeated statement to analyze repeated measures adjusting for time-dependent covariates (21). We also calculated a direct estimate of the mean within individual difference for the three measures of folate intake between the baseline and follow-up. The differences in the folate intake measures between baseline and follow-up were normally distributed; thus, no data transformations were required. The resulting mean differences were similar, but not identical to the difference of the mean intakes at follow-up and baseline because these latter means were based on transformed data.

Total folate, natural folate and folic acid intake were adjusted for age, gender, number of cigarettes smoked per day, body mass index and total energy intake. In addition to these factors, the mean differences of the three folate intake measures were also adjusted for their respective baseline intakes. We used the Tukey adjustment to account for multiple comparisons between means.

We also calculated the prevalence of total folate intake below the estimated average requirement (EAR) for folate (320 μg of DFE/d) (14), prevalence of folic acid intake above the UL for folate (1000 μg/d) (14) and the prevalence of both low cobalamin status (B-12 < 260 pmol/L or 350 pg/mL) (22) and folic acid intake above the UL (>1000 μg/d) in the exposed and nonexposed groups at baseline and follow-up. The prevalence of intakes below the EAR was adjusted by the variables mentioned previously. Because of the small numbers of subjects with folic acid intakes above the UL, we present only the crude prevalence for the intake measures based on the UL.

RESULTS

Characteristics of the population.

We estimated folate intake in a group of 692 nonusers and 239 users of supplements containing B-vitamins who were not exposed to fortification at follow-up (nonexposed group) and in another group of 389 nonusers and 160 users of B-vitamin supplements who were exposed to fortification at follow-up (exposed group) (Table 1). At the time of the 5th examination, the age range of the cohort was between 30 and 80 y and the mean age in each group was between 54 and 56 y. The proportion of women within nonusers and users of B-vitamin supplements was similar between the exposed and nonexposed group, although women were more likely to use B-vitamin supplements than men.

TABLE 1

Estimation of folate intake among individuals from the Framingham Offspring Cohort exposed and not exposed to folic acid fortification, according to folic acid supplement use¹

	Folic acid vitamin supplement use
	No		Yes
	Nonexposed	Exposed	Nonexposed	Exposed
n	692	389	239	160
Mean age at baseline, y (range)	54 (32–77)	56 (30–80)	55 (32–79)	56 (38–79)
% Female	47	48	57	58
Mean FA intake,2μg/d (95% CI)
Baseline3	39 (33–45)	30 (23–37)	409 (382–438)	418 (385–453)
Follow-up3	32 (28–37)	231 (215–248)*, †	413 (385–441)	634 (592–678)*, †
Difference	−12 (−22, −1)	190 (176, 204)*	19 (−19, 56)	219 (173, 264)*
Mean NF intake,2μg/d (95% CI)
Baseline	250 (245–256)	259 (252–266)	280 (269–291)	292 (279–306)
Follow-up	263 (257–268)†	260 (253–267)	287 (277–298)	292 (280–305)
Difference	21 (13, 28)	0.3 (−9, 10)*	13 (−2, 29)	−2 (−21, 17)
Mean total folate intake, μg DFE/d4 (95% CI)
Baseline	368 (354–383)	352 (333–371)	994 (945–1044)	1031 (972–1093)
Follow-up	360 (347–373)	665 (641–690)*, †	1013 (964–1063)	1384 (1314–1457)*, †
Difference	1 (−19, 21)	323 (296, 350)*	45 (−21, 111)	370 (289, 450)*
% < EAR for folate5 (95% CI)
Baseline	50.0 (46.7–53.4)	48.6 (44.2–53.1)	0.6 (0–1.9)	1.1 (0–2.6)
Follow-up	47.0 (44.1–49.9)	7.0 (3.1–10.9)*, †	1.0 (0.1–1.9)	0.3 (0–1.4)

1

Abbreviations: CI, confidence interval; DFE, dietary folate equivalent; EAR, estimated average requirement; NF, natural folate; FA, folic acid.

2

FA, NF and total folate means at baseline and follow-up are based on square root transformations. The difference represents the mean of the within-subject difference between follow-up and baseline and was not transformed. Therefore, the mean difference may not be equal to the difference of the transformed means. All means were adjusted for age, sex, number of cigarettes smoked/d, body mass index and energy intake at the appropriate examination; the mean difference was also adjusted for baseline values.

3

Baseline refers to 5th Framingham Offspring examination, which was completed before the start of FA fortification. Follow-up refers to the 6th examination, which started before fortification was implemented and extended over the time period during which fortification was implemented. Fortification was not in place during the 6th examination at the time the nonexposed group was examined but was in place at the time the exposed group was examined.

4

μg DFE = μg of natural folate + (1.7 × μg of FA).

5

320 μg DFE/d.

*

P < 0.001 for comparison with nonexposed group.

†

P < 0.001 for the comparison with baseline value.

Estimation of the effect of fortification on folic acid, natural folate and total folate intake.

The multivariable-adjusted folic acid, natural folate and total folate mean intakes are reported in Table 1 for both groups at baseline and follow-up. Among individuals who did not use B-vitamin supplements, folic acid, natural folate and total folate intakes did not differ between groups at baseline. Between baseline and follow-up, folic acid intake increased by a mean of 190 μg/d (P < 0.001), natural folate remained unchanged and total folate intake increased by a mean of 323 μg DFE/d (P < 0.001) among individuals in the exposed group. Among individuals in the nonexposed group, only natural folate increased significantly by a mean of 21 μg/d (P < 0.001). Among those who used B-vitamin supplements, folic acid, natural folate and total folate intakes were not different at baseline when the exposed and nonexposed groups were compared. Between baseline and follow-up, folic acid intake increased by a mean of 219 μg/d (P < 0.001), natural folate intake was unchanged and total folate intake increased by a mean of 370 μg DFE/d (P < 0.001) as a result of fortification among individuals in the exposed group.

Similar changes occurred in men and women. Among those who did not use supplements, the intake of folic acid after exposure to fortification increased 198 μg/d in men and 192 μg/d in women. The increase in folic acid intake associated with fortification was 272 and 182 μg/d, respectively, in men and women who used supplements.

Prevalence of individuals with total folate intake below the EAR for folate.

Among those who did not use B-vitamin supplements, the multivariable-adjusted prevalence of individuals with total folate intake below the EAR was the same at baseline in the nonexposed (50%) and the exposed groups (49%). At follow-up, among individuals in the exposed group, the prevalence of individuals with total folate intake below the EAR decreased to 7% (P < 0.001). Among those who used B-vitamin supplements, the prevalence of individuals with total folate intake below the EAR was ∼1% and was not different in the exposed and nonexposed groups at baseline or follow-up (Table 1).

The effect of fortification on the prevalence of folate intake below the EAR was similar in men and women. Among the 202 men who did not use supplements and were exposed to fortification at the 6th examination, 86 (43%) had intakes below the EAR at baseline, whereas only 12 (6%) had intakes below the EAR after exposure to fortification. Similarly, for the 187 women who did not use supplements and were exposed to fortification at the 6th examination, 102 (55%) and 16 (9%) had intakes below the EAR before and after exposure to fortification, respectively. Among those who used supplements and were exposed to fortification at the 6th examination, there were no men and only two women who had intakes below the EAR at baseline and no men or women with intakes this low after fortification. Among those not exposed to fortification, the prevalence of men and women with intakes below the EAR did not change between baseline and follow-up and was similar to the baseline prevalence for those exposed to fortification.

Prevalence of individuals with folic acid intake above the UL for folate.

Among those who did not use B-vitamin supplements, only two individuals at baseline and only one at follow-up had folic acid intake above the UL of 1000 μg/d. Among those who used B-vitamin supplements, the prevalence of individuals with folic acid intake above the UL was essentially the same at baseline in the nonexposed (2.1%) and exposed groups (1.3%), and did not change over time in the nonexposed group. However, at follow-up, the prevalence of individuals in the exposed group with folic acid intake above the UL increased from 1.3% (n = 2) to 11.3% (n = 18) (P < 0.001).

With low vitamin B-12 status defined as plasma B-12 levels <260 pmol/L (350 pg/mL), no individuals had low vitamin B-12 status and folic acid intakes >1000 μg/d at baseline. At follow-up, only three individuals had low B-12 status and folic acid intakes >1000 μg/d. All three were supplement users who were exposed to fortification (crude prevalence = 1.9%).

DISCUSSION

In this cohort, the introduction of folic acid fortification appeared to increase folic acid intake by approximately twice as much as expected on the basis of earlier dietary modeling. According to our estimations, the addition of folic acid to cereal-grain products resulted in a mean increase in folic acid intake of 219 and 190 μg/d in individuals who did and did not take supplements containing B-vitamins, respectively. Initial projections of the effect of folic acid fortification on folate intake (12,23) predicted an average increase of ∼80–130 μg folic acid/d for adults >19 y old. For these estimations, it was assumed that enriched cereal-grain products contained the amount of folic acid specified in the applicable regulations. Our estimation of intake was carried out using data from a study in which folate content was analyzed in >150 enriched cereal-grain products and other products fortified with folic acid (including breakfast cereals) using a microbiological assay with trienzyme digestion (13). For each product, the measured amount of total folate was compared with the amounts declared in the label and the amounts of folic acid required by regulation. In a considerable number of the food products analyzed, the measured amount of folate was appreciably higher than the folic acid levels required to be added by regulation (13). The folic acid fortification regulation provides for a single level of fortification for some of the products (enriched bread, rolls and buns and enriched flour) and for a range of levels of fortification for others (enriched farina, corn grits, corn meals, rice, macaroni and noodle products). When a single level is specified, the regulation allows for “reasonable overages within the limits of current good manufacturing practice” (11). This provision allows manufacturers of these products some flexibility in the amount of folic acid added to these products to ensure that the amount of folic acid required by the regulations and declared on the labels will be present throughout the shelf-life of the product. The overages observed largely explained our substantially higher estimations of the increase in folic acid intake resulting from the implementation of folic acid fortification.

We assessed inadequacy of total folate intake estimates using the EAR, which is the parameter suggested by the DRI committee to assess inadequacy of a particular nutrient intake in a group of individuals (14). The EAR for folate for men and women >19 y old is 320 μg DFE/d, and by definition is the amount of folate that is estimated to meet the requirements of 50% of healthy individuals in this age group (14,24). On the basis of the EAR, the prevalence of individuals with inadequate folate intake (folate intake <320 μg DFE/d) was very small after implementation of folic acid fortification, 7% among individuals who did not use B-vitamin supplements and essentially 0% among B-vitamin supplement users.

We also calculated the prevalence of individuals with folic acid intakes above the UL for folate. By definition “the UL for a nutrient is the highest level of daily nutrient intake that is likely to pose no risk of adverse health effects to almost all individuals in the general population” (14). The UL for adults was estimated to be 1000 μg of folic acid/d. It is based on folic acid, not total folate, because there was no evidence to indicate that excessive intakes of natural folate caused adverse health effects. The UL was determined on the basis of studies suggesting that large intakes of folic acid could trigger or aggravate neurologic damage in individuals with B-12 deficiency (14). The prevalence of individuals with folic acid intakes above the UL was notable only among individuals who used B-vitamin supplements and were exposed to folic acid fortification (11.3%). However, only three individuals in our cohort had both low plasma B-12 (<260 pmol/L) and estimated folic acid intake >1000 μg/d, and they were all supplement users exposed to fortification. Although our preliminary indirect assessment of risk based on high folic acid intakes in association with low vitamin B-12 status found few subjects with both conditions, there has been no direct assessment of risk associated with increased folic acid intake in older individuals, nor has there been any systematic examination of potential adverse effects on children. Lewis et al. (23) estimated the postfortification total folate intake in different population groups including children using data from two national food consumption surveys with corrections to reflect the required levels of folic acid added to foods. Their estimations, which relied on theoretical, not measured, folate values, suggested that ∼15–25% of children between the ages of 1 and 8 y could have intakes of folic acid that surpassed the UL of 300 μg folic acid/d (1–3 y) or 400 μg folic acid/d (4–8 y). The lack of information on actual folic acid intakes of children and potential for high folic acid intakes in both children and older adults indicates a critical need for further research.

There are several potential sources of imprecision in our estimations of folate intake. The FFQ does not provide a precise estimate of nutrient intake, but it is a relatively inexpensive and simple way in which to assess average long-term intake of a nutrient (18). The semiquantitative FFQ used for this study was validated in the past to assess folate intake, (18,19) but it has not yet been validated in the era of fortification. Our estimation of folic acid in fortified foods, based on the calculated difference between the new measured total folate value and the total folate value for each food in the existing food composition database, may lead to an overestimation of folic acid intake. The limitation of this approach is that the current folate values listed in the food composition database were measured using the traditional method in which the food sample was treated with folate conjugase, whereas the new measured total folate content in fortified foods was determined using a new method that uses trienzyme extraction; this method has been shown to result in higher folate values compared with the traditional approach (25). The traditional method may underestimate total folate content due to incomplete release of folate from the food matrix and incomplete cleavage of the polyglutamate chain (25). However, the values of natural folate in enriched cereal-grain products are very small relative to the folic acid added to these products, and even if there is an underestimation of the current food composition database values of folate in enriched cereal-grain products, it would have little overall effect on our estimations of folic acid in fortified foods. Our inability to identify the variety of mixed dishes and breaded foods using the FFQ may also lead to underestimation of folic acid.

Another potential limitation of the folate data derives from the fact that our values of measured folate were not based on a systematic study of the U.S. food supply. Although these products were selected on the basis of market data and represent the top-ranked products covered by the fortification regulations, they comprise a survey of a limited number of foods within enriched cereal-grain categories. Furthermore, these foods were collected in the Washington, DC metropolitan area. We have no comparable data on foods available in the Southern New England region during the period of this study.

The limitations of our estimates of the change in folate intake associated with fortification underscore the critical need for food composition tables that differentiate between natural folate present in foods and folic acid added to foods to allow assessment of total folate and folic acid intake in individuals and populations (15).

Although our findings must be viewed in light of these aforementioned limitations, this study represents the first attempt at assessing the effect of fortification on folate intake in the United States using data on measured folate content of fortified foods. Moreover, in spite of the limitations of our estimates, our results are very consistent with recent studies that evaluated the effect of folic acid fortification on biochemical measurements of folate status (17,26,27). These studies demonstrated increases in measures of folate status greater than anticipated on the basis of an average estimated increase of 100 μg folic acid/d. Our group recently showed that in the Framingham Offspring Cohort, individuals who were exposed to fortification had a mean RBC folate that was ∼295 nmol/L (130 μg/L) higher than the mean of individuals not exposed to fortification (28) and that fortification resulted in a mean plasma folate increase of 12.2 nmol/L (5.4 μg/L) (17). On the basis of a number of studies (29–32) that assessed the effect of supplemental folic acid on RBC folate concentrations, the observed difference in RBC folate between those exposed and not exposed to fortification would be explained by an additional folic acid intake of 148–260 μg/d. On the basis of a number of studies (31–35) that assessed the effect of supplemental folic acid on plasma folate concentrations, the observed increase in plasma folate due to fortification would be explained by an additional folic acid intake of 212–318 μg/d. Our results are also supported by those of Wald et al. (36), who recently suggested that fortification is providing an additional 200 μg/d of folic acid. They constructed a model to estimate a dose-response relation between folic acid intake and risk of an NTD-affected pregnancy. This model was based on data from studies that measured the effect of supplemental folic acid on blood folate levels and a study that addressed the relationship between blood folate levels and risk of NTD. Their summary of folic acid supplementation trials suggested that 200 μg/d additional folic acid would result in an 11 nmol/L (5 μg/L) increase in plasma folate among older individuals (36), which is the magnitude of the increase in plasma folate previously reported in the Framingham Offspring cohort (17). Furthermore, their model predicted that a 19% decrease in NTD prevalence, as reported in the United States after implementation of fortification (37), would be explained by an ∼200 μg/d increase in folic acid intake.

Our initial attempt to estimate natural folate and folic acid intakes in this era of fortification suggests that at current levels of fortification, the increased levels of folic acid associated with fortification were approximately twice as high as projected. The result of this increase appears to be that only a small percentage of the U.S. adult population has inadequate folate intake as assessed by the EAR. However, there remain many unanswered questions regarding the safety and importance of the additional folic acid in the American diet, which can not be adequately addressed without a national database of folic acid and total folate content of fortified foods. Although the purpose of this fortification was to assist women of childbearing age in increasing their folic acid intakes to help reduce NTD, there is probably no single level of fortification that will ensure that all women of childbearing age achieve recommended intakes of folic acid above the 400 μg/d and also ensure that the rest of the population have intakes that do not exceed the UL.

We thank Gail Rogers, for her help in the statistical analyses.

Abbreviations

 
• DFE

  dietary folate equivalent

 
• DRI

  dietary reference intake

 
• EAR

  estimated average requirement

 
• FDA

  Food and Drug Administration

 
• FFQ

  food-frequency questionnaire

 
• FHS

  Framingham Heart Study

 
• NTD

  neural tube defect

 
• UL

  tolerable upper intake level

FOOTNOTES

2

Supported in part by the U.S. Department of Agriculture under agreement No. 58–1950-9–001, by the National Heart, Lung and Blood Institute's Framingham Heart Study contract N01-HC-38038 and by National Institutes of Health grant 1R01 DK 56105–01.