The mean TFA intake in children and adults in the United States between 1999 and 2002 was 5.0–7.8 g/day (2.3–2.7 % of energy). In all cohorts, the range in TFA intake from the first quintile (1.3–2.0 g/day) to the fifth quintile (9.6–16.5 g/day) varied by approximately 5- to 8-fold. On a percentage-of-calories basis, TFA consumption ranged from 0.8–1.2 % (quintile 1) to 4.4–5.0 % of calories (quintile 5). For all population groups, there was remarkable variation in the range of TFA intake on a grams-per-day basis, as well as a percentage-of-energy basis in quintile 5. Variation was greatest in males aged 12–19 years, with intake that varied from 12 to 92 g/day in quintile 5. Collectively, these data illustrate that some individuals consume large quantities of TFA (in quintile 5), whereas there are others who eat little TFA (in quintile 1). Interestingly, as TFA intakes increased across quintiles, energy, total fat, and SFA intakes also increased in all populations. However, there were important differences in the magnitude of increases for each of these dietary constituents. Although there were noticeable increases in total fat and SFA intake across quintiles, as well as somewhat modest increases in energy intake across quintiles (specifically for the fatty acid data presented as percentages of energy in Table ), the proportional increase in TFA was dramatically greater.
Collectively, our results indicate a change in dietary patterns across quintiles. Specifically, TFA-containing foods were either added to the diet or substituted for non-TFA-containing foods. Given that the proportional increase in dietary TFA was greater than the increase in energy, it seems that substitution of TFA-containing foods was more common. Thus, a high TFA intake reflects a different dietary pattern compared with a lower TFA intake. In fact, starting with quintile 2, it appears that the dietary patterns are shifting because of increases in the proportions of TFA, SFA, and total fat (as a percentage of energy) relative to the increase in energy intake. However, although all population groups showed similar changes in fatty acids and energy across quintiles, it can be concluded that dietary patterns within a quintile are likely comparable among the different population groups.
It is telling that the results of this study show that the population evaluated in 1999–2002 consumed similar amounts of TFA compared to the U.S. population surveyed in 1989–1991 [11
]. In fact, the mean and median TFA intakes are similar in both studies. However, what is striking between the two studies is the increase in TFA intake in the 90th percentile. As Table shows, there were approximately 20 and 27 % higher TFA intakes in men and women, respectively, in the 1999–2002 database versus the earlier survey. Collectively, this finding suggests that there is a relatively large cohort of individuals in the population who have made changes in their diet in a manner that is incongruent with current dietary recommendations. It is important to appreciate that the large gap between the 99.9th and 100th percentiles of TFA intake (34 vs. 92 g/day) most likely represents only a very small group of high-TFA consumers, which would be expected to slightly skew the results for this cohort. Nonetheless, these findings indicate that there are individuals who follow extreme dietary practices, putting them at high risk for chronic disease and malnutrition.
The food sources of dietary TFA are similar among the different population groups studied. The major TFA sources were cakes, cookies, pies, and pastries, as well as yeast breads, French fries, grains and ethnic dishes, and tortilla chips. It is important to note that many food sources of TFA are also major contributors of SFA [8
], such as grain-based desserts, savory snacks, ethnic dishes, and French fries/fried potatoes. Many of these foods are typically classified as discretionary calories and consequently should be limited in the diet. Reducing these foods would not only decrease TFA but also SFA and excess calories.
Our analysis provides useful information about TFA intake and food sources in the United States prior to the onset of legislative action intended to decrease TFA in the food supply. The data presented here are important as a benchmark to track changes in TFA intake in the future as the result of these sweeping legislative mandates to decrease TFA in the food supply. It will be important to monitor changes in fatty acid intake in the population because of the concerted effort to remove TFA from the food supply. Much progress has occurred in food science and lipid chemistry to appreciably decrease industrially produced TFA in the food supply in recent years. In fact, a new study reported an average population intake of 1.3 g/day of industrially produced TFA using analysis of 2003–2006 NHANES data [15
]. In many instances, oils high in MUFA are replacing conventional fats that are high in TFA (e.g., for liquid fat food applications), whereas in other situations, fat sources of SFA are replacing solid fats rich in TFA (e.g., for solid fat food applications). One caveat to be mindful of is that efforts to decrease TFA should not result in increases in SFA intake in the population. Consistent with this, the American Heart Association [16
] has recommended that TFA and SFA in unmodified foods not be greater than total SFA in modified foods. It is clear that there are currently countless fats that differ in their fatty acid profiles that could be substituted for TFA. It will be important that fats devoid of TFA be selected to achieve current dietary fatty acid guidelines and, thereby, realize a public health benefit. This has been reported in a recent modeling exercise by Lefevre et al. [17
]. The last point that must be emphasized is to acknowledge that many of the foods that deliver industrially produced TFA are “extras in the diet,” and even if they are modified to have a fatty acid profile that is consistent with current dietary recommendations, their intake should be limited within the context of a healthy diet that meets current food-based and nutrient recommendations. Limiting intake of these fatty acid-modified foods will also help control calories, which is a pressing societal need.
A limitation of this study is that the analysis of TFA intake was based on NHANES 1999–2002 data. The TFA data were derived from the 1995 Trans
Fatty Acid Database and the USDA Nutrient Database for Standard Reference (Release 15). Consequently, changes that were made in foods in the marketplace (with respect to TFA content) between 1993 (when the TFA analyses of foods were completed) and 1999–2002 are not reflected in this analysis. This is a well-recognized limitation of the available nutrient databases and underscores the importance of continuously updating them. Another potential limitation is the use of two 24-h recalls. According to Allison et al. [11
], 3 days of intake data provide a better assessment of usual intake. Moreover, the present study only evaluated industrial TFA and not other sources of TFA, which could have important health implications [18
]. Despite these limitations, however, this study provides useful information about TFA intake in the population and important food sources prior to collective efforts by the food industry to decrease TFA in the food supply.
In summary, in all population groups studied, TFA intake between 1999 and 2002 was 5.0–7.8 g/day (2.3–2.7 % of energy). Much of the TFA consumed was derived from energy-dense and nutrient-poor foods such as cakes, cookies, pies, pastries, and savory snacks (i.e., chips). Strikingly, there was a population, albeit small, with an extremely high intake of TFA (males aged 12–19 years in the fifth quintile, who consumed 11.8–92.4 g/day) as well as SFA (and also energy). Obviously, this reflects very poor dietary practices. In fact, as TFA intake increased in the population groups studied, there was a trend toward poorer dietary practices. Ongoing nutrition intervention efforts must be directed at decreasing TFA and also promoting healthy dietary patterns.