Intake of Added Sugars and Selected Nutrients in the United States, National Health and Nutrition Examination Survey (NHANES) 2003—2006

Abstract

In the Institute of Medicine (IOM) macronutrient report the Committee recommended a maximal intake of ≤ 25% of energy from added sugars. The primary objectives of this study were to utilize National Health and Nutrition Examination Survey (NHANES) to update the reference table data on intake of added sugars from the IOM report and compute food sources of added sugars. We combined data from NHANES with the United States Department of Agriculture (USDA) MyPyramid Equivalents Database (MPED) and calculated individual added sugars intake as percent of total energy then classified individuals into 8 added sugars percent energy categories, calculated usual intake with the National Cancer Institute (NCI) method, and compared intakes to the Dietary Reference Intakes (DRIs). Nutrients at most risk for inadequacy based on the Estimated Average Requirements (EARs) were vitamins E, A, C, and magnesium. Nutrient intake was less with each 5% increase in added sugars intake above 5–10%. Thirteen percent of the population had added sugars intake > 25%. The mean g-eq added sugars intake of 83.1 g-eq/day and added sugars food sources were comparable to the mid-1990s. Higher added sugars intakes were associated with higher proportions of individuals with nutrient intakes below the EAR, but the overall high calorie and the low quality of the U.S. diet remained the predominant issue. With over 80% of the population at risk for select nutrient inadequacy, guidance may need to focus on targeted healthful diet communication to reach the highest risk demographic groups for specific life stage nutrient inadequacies.

Keywords:

• NHANES
• added sugars
• total energy intake
• usual intake
• NCI method
• food sources

INTRODUCTION

The Dietary Guidelines for Americans (DHHS/USDA, 2005) provide recommendations for diet and exercise for maintaining good health. These guidelines are based on the latest scientific information on healthful diet, exercise, and lifestyle patterns in the United States (King, 2007; 2005 Dietary Guidelines Advisory Committee, 2004). Since the first publication in 1980, the guidelines have included advice for the U.S. population to moderate total sugar intake cf (USDA/DHEW, 1980; USDA/DHHS, 1985; USDA and DHHS, 1990; USDA and DHHS, 1995; Davis and Saltos, 1996; Guthrie and Morton, 2000; USDA/USDHHS, 2000; DHHS/USDA, 2005).

In 2002, the Institute of Medicine (IOM) released the Dietary Reference Intake (DRI) report on macronutrients as part of its series of evidence-based nutrient intake recommendations for the nation (IOM, 2002b). While not all foods that contain high levels of added sugars are necessarily nutrient-poor, a consideration in this report was whether a high level of added sugars in energy-dense, nutrient-poor (EDNP) foods in the diet contributed a risk for low micronutrient intake and hence reduced diet quality. Such foods could be compared to the sources of naturally occurring sugars as found in fruits and dairy products, which contain higher levels of micronutrients. Data from the third National Health and Nutrition Examination Survey (NHANES III: 1988–1994) indicated that EDNP foods provided approximately 27% of energy intake of the diet in the U.S. at the expense of nutrient-dense foods (Kant, 2000).

Without consideration of added sugars, diets in the U.S. are low in specific nutrients. A recent report on national estimates of the usual intake of 24 nutrients and dietary components from food based on the NHANES 2001–2002 indicated that a large percent of the U.S. population across 17 age/gender groups was at risk for low intakes of vitamins A, E, and C and magnesium when compared to the IOM Estimated Average Requirements (EAR) for those nutrients (Moshfegh et al., 2005).

The IOM macronutrient report did not recommend a tolerable upper intake level (UL) for total or added sugars, but did suggest a maximal intake level of 25% or less of energy from added sugars for both adults and children. Part of the rationale for this recommendation was concern about low micronutrient intake of persons whose diet exceeded 25% of energy from added sugars. The suggestion was based on the published literature and data on median intakes of selected micronutrients at 5% increments of added sugars intake (from 0 to >35%), developed from NHANES III data, and referenced in the report as Appendix Table J.

Additional potential relationships reviewed in the IOM report were the role of total and added sugars in dental caries, behavior, cancer, risk of obesity, and risk of hyperlipidemia. Macronutrient distribution in general was considered in terms of the maintenance of body weight, and the risk of coronary heart disease (CHD), hyperinsulinemia, glucose tolerance, and Type 2 diabetes.

Since the publication of the IOM macronutrient report, the scientific discussion about the role of energy, added sugars, and food sources of sugar, including the role of soft drink consumption, has continued (Ballew et al., 2000; Krebs-Smith, 2001; Nielsen et al., 2002; Bray et al., 2004; Storey et al., 2004; Bachman et al., 2006; Malik et al., 2006; Vartanian et al., 2007; Keller et al., 2009). Part of this discussion has included questions about the methodology used to generate the IOM report Appendix Table J results. The original IOM report Appendix Table J approach was criticized on two grounds (Forshee and Storey, 2004; Forshee and Storey, 2005). First, the use of a ratio variable (added sugars calories over total calories) as the basis for establishing analytic categories was questioned because of inherent statistical complications involved in analyzing ratio variables, and, in this specific application, because of possible dependency between the numerator and the denominator. Second, these authors argued that the model used to estimate usual daily intake of nutrients for each of the added sugars categories should control for total energy intake, since nutrient intake may be higher among those who consume more calories. To the extent that the total energy intake is correlated with the percent energy intake from added sugars, the observed relationship between nutrient intake and percent energy from added sugars could be confounded.

Others responded, rejecting the criticism based on the use of a ratio variable, but agreeing that the IOM report Appendix Table J analysis should have controlled for total energy intake when estimating usual daily intake of nutrients (Barr and Johnson, 2005). In addition, these authors stated that a broader understanding of both the food sources of added sugars and sub-population composition was needed. These criticisms, which questioned the micronutrient estimates referenced by the IOM report in their Appendix Table J, therefore had implications for the IOM committee recommendation of limiting added sugars intake to less than 25% of daily energy.

The purpose of the present study was to update and expand the estimates of selected nutrient intake as related to dietary added sugars. We used the most recent NHANES data and corresponding USDA MyPyramid Equivalent Databases, and assessed the intake of a selected larger group of nutrients, including fiber, than were analyzed in the original IOM report Appendix Table J. In addition, we used the newly-developed National Cancer Institute (NCI) usual intake estimation approach (Tooze et al., 2006), and controlled for total energy intake in our analyses to facilitate comparisons across subgroups. We also included demographic information on the NHANES population as it related to added sugars consumption as a percent of total energy intake, and describe differences in added sugars intake by food source across these groups. Due to the extensive differences in methodology we do not believe that the numerical data presented in this paper can be compared directly with the information presented in Appendix Table J in the IOM report on macronutrients (IOM, 2002b). Instead, we contend that the information in this paper must be viewed as a new assessment of the question.

METHODS

Sample

Our analyses employed two days of 24-hour dietary recall data from the National Health and Nutrition Examination Survey (NHANES), What We Eat in America 2003–2006. We restricted our sample to children and adults as defined by the DRI life stage groups, i.e. individuals aged 4 years or older. Our sample was further restricted to individuals with reliable recall status, excluding fasters, pregnant women, and lactating women. The final analytic sample included 29,099 days of recall data from 15,189 individual respondents. ¹ We also analyzed data from the NHANES 2001–2002 for methodological comparison.

BMI (kg/m²) for this study was calculated from the measured height and weight collected in the NHANES mobile examination center. Respondents aged 21 or older were classified as underweight if BMI was less than 18.5, normal weight if BMI was 18.5 or above but below 25, overweight if BMI was 25 or above but below 30, and obese if BMI was 30 or above. Respondents aged 20 or younger were classified by their BMI age/sex percentile ranking relative to 2000 CDC Growth Reference values—underweight if BMI was below the 5th percentile, normal weight if BMI was at or above the 5th percentile but below the 85th percentile, overweight if BMI was at or above the 85th percentile but below the 95th percentile, and obese if BMI was at or above the 95th percentile.

All analyses reported here incorporated NHANES dietary recall sample weights and appropriate stratification and clustering methodology to account for the complex NHANES sampling design (CDC, 2005). Therefore, these results may be considered representative of the U.S. population.

Classifying Individuals by Added Sugars Intake

Using the USDA MyPyramid Equivalents Database (MPED) 2.0 (Bowman et al., 2008), which was developed for the 2003–2004 NHANES, we calculated the added sugars content in each individual food reported on the NHANES recalls. For 68 new foods consumed in 2005–2006, a nutritionist on our team (PC) developed added sugars estimates based on added sugars content in similar foods already included in the database. This step was necessary because at the time of our analyses, USDA had not yet released MPED 3.0, which is being developed for the 2005–2006 NHANES.

MPED servings are expressed in teaspoons of sugar (Bowman et al., 2008). “Added sugars include all sugars used as ingredients in processed and prepared foods such as breads, cakes, soft drinks, jams, chocolates, and ice cream, and sugars eaten separately or added to foods at the table… Added sugars do not include naturally occurring sugars such as lactose in milk or fructose in fruit, unless the sugar is added to the food item” (USDA Center for Nutrition Policy and Promotion, 2005; Bowman et al., 2008). In the MPED that we used for this study “Added sugars are defined as white sugar, brown sugar, raw sugar, corn syrup, corn syrup solids, high fructose corn syrup, maple syrup, pancake syrup, fructose sweetener, liquid fructose, honey, molasses, dextrose, and dextrin” (Bowman et al., 2008). The 1986 report of the Food and Drug Administration (FDA) Sugar Task Force defined added sugar as the total of monosaccharides and disaccharides (Glinsmann et al., 1986). Because our analyses are based on the MPED data, we included other saccharides in addition to the monosaccharides and disaccharides specified in the FDA definition (Guthrie and Morton, 2000). However, for this paper we have chosen to use the term “added sugars” to represent the full range of sugars encompassed by the definition.

From these food-level added sugars estimates, we calculated energy from added sugars as a percent of total energy for each individual in the NHANES sample based on foods the individuals reported that they consumed on each recall day. Similar to the earlier IOM report Appendix Table J analyses, we placed each individual into one of eight added sugars categories from 0 to >35% energy intake from added sugars. We present the distribution of the NHANES sample by these eight added sugars categories across 13 DRI life stage groups (excluding lactation and pregnancy) with selected characteristics for each added sugars category in Table 1. However, because of small sample sizes in the lowest and highest added sugars categories, in the remainder of this paper we present results for three broader composite added sugars categories for the 13 DRI life stage groups—individuals ingesting between 0 and 15% of energy from added sugars, more than 15% but less than or equal to 25% of energy from added sugars, and over 25% of energy from added sugars. ²

Table 1 Sample Size, Weighted Population Estimates, Percentage, Estimated Daily Energy Intake, Estimated Daily Intake of Added Sugars, Dietary Reference Intake (DRI) Life Stage Groups ^a , and Selected Demographic Characteristics by Range of Percent of Estimated Daily Intake from Added Sugars, National Health and Nutrition Examination Survey (NHANES) 2003–2006

	Categories of Individuals Based on Percent of Energy Intake from Added Sugar^b
Characteristic^c	Full Sample^c	0 ⩽ x ⩽ 5%	5 < x ⩽ 10%	10 < x ⩽ 15%	15 < x ⩽ 20%	20 < x ⩽ 25%	25 < x ⩽ 30%	30< x ⩽ 35%	>35%
N	15,190	1387	2877	3666	3311	2055	1078	432	384
Weighted N	287,845,042	33,743,743	60,083,772	68,584,551	55,570,985	33,778,579	18,911,454	8,754,407	8,417,551
Percent	100.0	11.7	20.9	23.8	19.3	11.7	6.6	3.0	2.9
Estimated Total Daily Energy Intake, Kcal: Median (SE)	2118 (13.1)	2028 (17.4)	2063 (15.4)	2104 (12.9)	2138 (14.1)	2177 (15.2)	2221 (19.7)	2261 (22.49)	2298 (28.3)
Estimated Daily Added Sugar Intake, With Total Energy Intake as a Covariate in the Analysis Gram Equivalents: Mean (SE)^d	82.9 (0.06)	24.9 (0.06)	44.9 (0.07)	66.9 (0.07)	91.5 (0.08)	127.1 (0.13)	160.3 (0.21)	197.6 (0.40)	232.8 (0.50)
Mean Age: yr (SE) ^e	38.6 (0.5)	47.2 (0.9)	44.5 (0.8)	38.4 (0.7)	34.8 (0.7)	32.2 (0.6)	30.9 (0.8)	32.6 (1.1)	33.9 (1.2)
Gender: % (SE)
Male	49.4 (0.6)	11.1 (0.8)	20.5 (0.9)	23.6 (0.9)	20.1 (0.8)	12.6 (0.7)	6.8 (0.4)	3.0 (0.4)	2.4 (0.3)
Female	50.6 (0.6)	12.0 (1.0)	22.6 (0.9)	24.5 (0.8)	18.8 (0.6)	10.7 (0.6)	6.2 (0.5)	2.6 (0.3)	2.6 (0.4)
DRI Group: % (SE)
Children aged 4–8	7.5 (0.3)	3.3 (0.7)	15.4 (1.4)	31.4 (2.0)	24.8 (1.7)	15.1 (1.3)	7.5 (1.4)	1.9 (0.5)	0.7 (0.4)
Males 9–13 yrs	4.1 (0.2)	2.3 (0.7)	10.1 (1.5)	24.8 (2.0)	31.2 (2.8)	17.2 (2.5)	9.7 (2.0)	2.9 (1.0)	1.8 (0.7)
Females 9–13 yrs	3.8 (0.2)	4.3 (1.2)	16.0 (1.8)	25.4 (2.3)	27.8 (2.3)	15.1 (1.7)	8.6 (1.0)	1.6 (0.7)	1.2 (0.4)
Males 14–18 yrs	4.1 (0.3)	2.7 (0.6)	10.6 (1.6)	19.2 (2.1)	27.8 (2.3)	22.7 (2.3)	9.4 (1.5)	4.3 (0.7)	3.3 (0.8)
Females 14–18 yrs	3.8 (0.3)	3.4 (0.8)	13.1 (1.7)	20.0 (1.7)	25.7 (1.8)	17.5 (1.8)	12.6 (1.7)	3.7 (1.0)	4.0 (1.1)
Males 19–30 yrs	8.1 (0.5)	10.1 (1.5)	15.9 (1.7)	22.7 (2.1)	18.4 (1.7)	14.9 (1.5)	8.6 (1.3)	5.3 (1.2)	4.2 (1.0)
Females 19–30 yrs	7.2 (0.4)	9.8 (1.7)	16.2 (2.2)	22.9 (1.8)	19.3 (2.0)	12.5 (1.5)	9.4 (1.7)	5.0 (1.0)	4.9 (1.1)
Males 31–50 yrs	15.0 (0.5)	13.4 (1.5)	21.7 (1.3)	23.2 (1.6)	18.6 (1.2)	11.1 (1.3)	6.7 (0.7)	2.4 (0.5)	2.9 (0.6)
Females 31–50 yrs	15.1 (0.5)	14.3 (1.6)	23.4 (1.7)	23.4 (1.8)	15.6 (1.5)	11.4 (1.2)	5.5 (0.7)	3.1 (0.6)	3.4 (0.7)
Males 51–70 yrs	10.7 (0.5)	17.0 (1.4)	28.6 (1.7)	22.8 (1.7)	16.4 (1.6)	7.7 (1.1)	4.2 (1.1)	2.4 (0.6)	1.0 (0.3)
Females 51–70 yrs	12.3 (0.6)	18.0 (1.6)	30.2 (1.7)	24.0 (1.8)	15.2 (1.2)	5.8 (1.0)	3.9 (0.7)	1.4 (0.4)	1.5 (0.5)
Males >70 yrs	3.5 (0.3)	15.7 (1.7)	31.2 (2.3)	26.1 (1.6)	16.4 (1.5)	6.4 (0.8)	2.2 (0.7)	1.1 (0.5)	0.9 (0.5)
Females >70 yrs	4.6 (0.3)	12.6 (1.6)	28.3 (2.2)	27.9 (2.0)	19.2 (2.1)	7.7 (1.2)	2.2 (0.7)	0.8 (0.4)	1.3 (0.6)
Race/ethnicity: % (SE)
Non-Hispanic White	71.6 (2.3)	12.6 (0.8)	22.3 (0.9)	23.7 (0.9)	18.6 (0.7)	11.1 (0.6)	6.3 (0.5)	2.9 (0.4)	2.4 (0.3)
Non-Hispanic Black	11.8 (1.4)	7.2 (0.9)	15.3 (1.2)	24.5 (0.8)	23.2 (0.9)	14.7 (0.8)	8.7 (0.6)	3.2 (0.4)	3.2 (0.6)
Hispanic^f	11.3 (1.3)	8.3 (0.9)	21.0 (1.4)	27.4 (1.5)	20.8 (1.0)	13.5 (1.1)	5.6 (0.6)	2.2 (0.4)	1.2 (0.4)
Other Race/Ethnicity^f	5.3 (0.6)	15.4 (2.5)	26.6 (3.2)	19.6 (2.3)	19.6 (2.4)	7.8 (1.2)	5.1 (1.3)	1.5 (0.5)	4.4 (1.7)
Poverty-Income Ratio (PIR): % (SE)
Below Poverty Line	13.4 (0.8)	6.8 (0.9)	17.4 (1.6)	23.5 (1.3)	21.1 (1.0)	13.7 (0.8)	8.9 (1.1)	4.7 (0.9)	3.9 (0.6)
At/Above Poverty Line	37.5 (1.3)	9.7 (0.9)	21.0 (0.8)	23.4 (0.9)	20.8 (0.8)	12.3 (0.7)	6.7 (0.5)	2.8 (0.5)	3.3 (0.5)
Above 3x Poverty Line	49.1 (1.7)	14.4 (0.9)	23.2 (1.1)	24.6 (0.9)	18.0 (0.9)	10.5 (0.5)	5.6 (0.5)	2.2 (0.3)	1.4 (0.2)
Mean Body Mass Index (BMI):	26.6 (0.2)	28.9 (0.4)	27.4 (0.2)	26.1 (0.3)	25.5 (0.2)	25.6 (0.3)	26.1 (0.3)	25.5 (0.5)	28.1 (0.7)
BMI Weight Categories: % (SE)^g
Underweight	1.6 (0.1)	11.5 (2.9)	23.2 (3.4)	17.1 (3.5)	20.8 (2.8)	12.1 (2.6)	7.1 (2.2)	3.5 (1.6)	4.8 (1.8)
Normal weight	39.3 (1.0)	8.6 (0.7)	19.7 (1.0)	25.5 (1.2)	21.3 (0.9)	12.5 (0.6)	6.8 (0.5)	3.4 (0.5)	2.2 (0.4)
Overweight	29.5 (0.7)	12.8 (0.9)	22.5 (0.9)	24.3 (1.1)	18.2 (1.0)	11.1 (0.6)	6.5 (0.6)	2.7 (0.4)	2.0 (0.5)
Obese	29.6 (1.0)	14.5 (1.1)	22.9 (1.3)	22.1 (0.7)	18.2 (0.9)	11.0 (0.8)	6.0 (0.5)	2.0 (0.3)	3.2(0.5)
^aDietary Reference Intake Life Stage groups [Institute of Medicine (IOM), 1997].
^bOur analyses employed two days of 24-hour dietary recall data from the National Health and Nutrition Examination Survey (NHANES), What We Eat in America 2003–2006. We restricted our sample to children and adults as defined by the DRIs as aged 4 years or older. Our sample was further restricted to individuals with reliable recall status, excluding fasters, pregnant women, and lactating women. The final analytic sample included 29,099 days of recall data from 15,189 individual respondents. For 2,539 individuals in the sample, only one 24-hour recall was reported.
^cPercentages for characteristics are indicated in this table as follows: in the full sample column the percent for each level of the characteristic totals to 100%; for rows of levels within a characteristic each specific row totals to 100%. For example, while the full sample was comprised of 49.5% male, of the total 100% of these males 23.6 % ingested >10 but ⩽15 % of their energy from added sugar.
^dOne Gram–equivalent (g-eq) equals an amount of added sugar comparable to 1 g sucrose in carbohydrate content. Added sugar gram-equivalents were calculated based on two-day mean intake from MyPyramid Equivalents Database servings.
^eNHANES top-codes age at 85 years, therefore the mean age estimate is biased slightly downwards.
^fHispanic includes Mexican-Americans and Other Hispanics. Other race ethnicity includes 1) individuals whose reported racial or ethnic identity was not Mexican American, Other Hispanic, Non-Hispanic White, or Non-Hispanic Black and 2) multiracial persons.
^gBMI (kg/m^2) for this study was calculated from the measured height and weight collected in the NHANES exam unit. Respondents aged 21 or older were classified as underweight if BMI was less than 18.5, normal weight if BMI was 18.5 or above but below 25, overweight if BMI was 25 or above but below 30, and obese if BMI was 30 or above. Respondents aged 20 or younger were classified by their BMI age/sex percentile ranking relative to 2000 CDC Growth Reference values: underweight if BMI was below the 5th percentile, normal weight if BMI was at or above the 5th percentile but below the 85th percentile, overweight if BMI was at or above the 85th percentile but below the 95th percentile, and obese if BMI was at or above the 95th percentile.

For those individuals with two complete days of dietary recall data, we used the two-day mean intake estimate; for individuals with only one day of dietary recall data, the one-day estimate was used. ³ Incorporating the second day of dietary recall data wherever possible reduces the observed variation stemming from day-to-day variation in individual diets; however, it is likely that the variance of the two-day mean distribution remains larger than that of the true usual intake distribution (Dodd et al., 2006), implying that some individuals are likely to have been “misclassified” based on their two-day mean added sugars intake relative to their actual usual added sugars intake.

An alternative approach to address this potential “misclassification” would have been to estimate individual-level added sugars intake as a percent of total energy via a regression-based model, incorporating covariates such as food frequency variables and demographic characteristics, and using the predicted results to assign individuals to intake categories. We tested this approach, but because these specifications had extremely poor model fit, we abandoned it in favor of the two-day mean intake method. ⁴

The NCI Usual Intake Method

In the next phase of our analysis, we generated the usual dietary intake estimates for nutrients and for added sugars by food source, employing methodology recently developed by the National Cancer Institute (NCI). All analyses were completed in SAS 9.2 using NCI-supplied macros MIXTRAN and DISTRIB (Version 1.1) to estimate the usual intakes. ⁵ For nutrient intake analyses, we used the NHANES-provided Day 1 dietary recall weights. For added sugars intake analyses by food group, for which food frequency questionnaire (FFQ) variables were included as covariates (see below), we used the appropriate FFQ weights.

For this study, we used the NCI method and assessed usual dietary intakes, or long-term average intakes, in order to characterize nutritional adequacy for different population subgroups. Since not all foods are consumed every day, statistical methods must specifically account for the observed zero-consumption days. Additionally, the usual intake estimation procedures separately model within-person variation (i.e., day-to-day differences in dietary consumption for a single individual) and between-person variation (i.e., differences in dietary consumption across individuals). The NCI method models usual intake as the product of the probability of dietary consumption on a given day and the average amount of dietary consumption consumed per consumption day. See (Tooze et al., 2006) for a detailed overview of the NCI method.

While this study employed the recently-developed NCI method for estimating the usual intake, the IOM report Appendix Table J analysis used the Iowa State University (ISU) method, which has been the previous standard for estimating usual intake (Nusser et al., 1996; Carriquiry, 2003). Both the ISU and NCI methods take into account reported zero-consumption days and reported consumption-day amounts that are positively skewed, and both methods distinguish between within-person and between-person variation in consumption. However, we chose to use the new NCI method because it additionally allows for correlation between the amount and the frequency of consumption, and permits the incorporation of covariates such as information from an FFQ, which in some cases can improve the power to detect relationships between dietary intake and other variables (Subar et al., 2006).

As described below, we exploited the latter capability of the NCI method by incorporating information from the NHANES 2003–2006 FFQ to generate estimates of usual intake of added sugars by food category source. We additionally incorporated total energy intake as a covariate in both our nutrient and added sugars intake models, in order to control for differences in usual intake that may be driven by variation in total energy intake. Incorporation of the total energy intake as a covariate allowed us to control for the total energy intake for each individual, which was a concern about the earlier IOM report Appendix Table J analysis (Forshee and Storey, 2004). The total energy intake is a concern because individuals who ingest more total calories could have higher intake of specific nutrients than individuals who ingest less total energy. By controlling for total energy intake in the analysis, we enabled appropriate comparisons of the nutrient data across subgroups whose average energy intake may differ. That is, any differences across subgroups that persist after controlling for total energy may be attributed to factors other than differences in total energy intake across those subgroups.

Finally, unlike the ISU method, the NCI method allows for efficient estimation of the usual intake for subgroups. Instead of stratifying the sample by subpopulation and estimating the usual intake separately for each subgroup (added sugars and DRI life stage categories), covariates defining subgroups were included in the NCI model, such that variance components were estimated simultaneously for the full sample, and only covariate values differed. Because we produced estimates for relatively small categories defined by percent energy intake from added sugars, further subdivided by life stage groups to correspond to DRI recommendation groupings, the efficiency gains from this capability were likely to be relatively substantial in this application.

The NCI method for estimating the usual intake was released in 2008, while the ISU method has been used since the 1990s (Nusser et al., 1996; Carriquiry, 2003; Carriquiry and Camano-Garcia, 2006). For this study, we also applied the NCI method to data from NHANES 2001–2002. We then compared the resulting estimates of the usual nutrient intake across the 13 DRI life stage groups with estimates from a previously published study that were obtained using the ISU method (Moshfegh et al., 2005). We made this comparison as a check on our analyses because the NCI method and our use of it were new. Although one would expect some minor differences between our results and those of the previous study due to differences in methodology, we conducted these analyses primarily to check for very large discrepancies between the published results and our own that might point to concerns about our use of the new NCI methodology. Because a second day of recall data was collected for only 2002 NHANES respondents, and because these data (though used in the prior study) were unavailable for public use, our own 2001–2002 intake estimates were based on only a single day of 2001–2002 NHANES recall data. We applied the second-day recall data from NHANES 2003–2006 to estimate the distribution of intakes; our estimated standard errors and confidence intervals are therefore likely to differ slightly from the published data that used the actual 2002 second-day subsample. The results of these supplementary analyses, in which, respectively, we did not and did control for total energy intake, are included in Appendix Tables 1 and 2, so that readers may directly compare our NCI method estimates with prior published estimates for 2001–2002. In general, our results do not differ greatly in comparison to the earlier (Moshfegh et al., 2005) estimates.

Model Specification

In this analysis, we produced usual intake estimates for two types of dietary components—1) nutrients and 2) added sugars by food source. Nutrients such as calcium, iron, or vitamin C are consumed by nearly every individual in the population on a daily basis. In contrast, foods containing added sugars may be consumed on a more episodic basis. For example, an individual might drink regular soda only when dining away from home. Estimating the usual intake of episodically-consumed dietary components such as added sugars required a slightly different methodological approach than estimating the usual intake of dietary components like nutrients that were consumed nearly every day.

We used the NCI method to estimate usual intake of calcium, magnesium, vitamin A, vitamin E, iron, zinc, vitamin C, vitamin K, folate, phosphorus, sodium, potassium, total choline, and fiber. ⁶ Because these nutrients are regularly consumed by nearly everyone in the sample on a daily basis, it is unnecessary to separately model the probability of their consumption. For nutrients, we therefore employed a one-part, or “amount-only,” version of the NCI method, in which only the amount of consumption was estimated.

Food Sources of Added Sugars

For this study we additionally estimated the usual intake of added sugars by food source for the following food categories based on USDA Food Survey Research Group codes used to classify NHANES foods—milk and milk products; eggs, meat, poultry, fish, and mixtures; legumes, nuts, and seeds; grain products; fruits; vegetables; fats, oils, and salad dressings; and sugar, sweets, and beverages. To facilitate a comparison with earlier work on added sugars food sources (Guthrie and Morton, 2000), we further subdivided the grain products into breakfast cereals, sweetened grains, and other grains, and the sugar, sweets, and beverages grouping into sugars/sweets, regular soft drinks, regular fruitades/fruit drinks, low-calorie beverages, alcoholic beverages, and other non-alcoholic beverages. Unlike nutrients, not all individuals consume added sugars from these sources on a daily basis. For these usual intake estimates, we therefore employed a two-part model, in which both probability and the amount of consumption were estimated.

Model Covariates

Our analyses exploited the ability of the NCI method to incorporate covariates into usual intake estimates. In both our nutrient models and our added sugars by food source models, we included an indicator to account for differences in weekend consumption, and another indicator to control for differences in total energy intake. In our added sugars by food source models, we additionally incorporated covariates from the NHANES 2003–2006 FFQ.

Weekend Effects

Both the nutrient models and the added sugars by food source models incorporated a covariate indicating whether the recall day fell on a Friday through Sunday, since weekend consumption may vary relative to consumption during the week (Moshfegh et al., 2005). For the nutrient models, this covariate was used as a predictor in both the probability and the amount model specifications.

Total Energy

Due to concerns about limitations on the interpretation of the nutrient results in the IOM macronutrient report Appendix Table J, which did not control for total energy intake (Forshee and Storey, 2004), we included the total energy intake as a covariate in both the nutrient and the added sugars by food source models. This permitted us to control for the possibility that differences in intake across added sugars categories were driven by differences in total energy consumption across these categories rather than by differences related to added sugars consumption. It should be noted that the results from our models are not therefore directly comparable to results from models that do not similarly control for total energy intake, such as the IOM report Appendix Table J results, the Bowman added sugars groupings (Bowman 1999) or the Guthrie and Morton (2000) added sugars by food source estimates. Similar to the weekend indicator, total energy was included as a covariate in both the probability and amount specifications in the two-part added sugars by food source models. For comparison purposes, we additionally ran a second group of models that did not control for total energy, in order to determine how the inclusion of this covariate influenced our estimates.

The usual intake estimates from our energy-adjusted specifications may be interpreted as regression-adjusted medians, evaluated at mean total energy for the relevant population subgroup. In this paper, when making comparisons across subgroups we use the energy-adjusted results, because the observed unadjusted differences in nutrient intake for subgroups defined by age, gender, or average added sugars intake may be attributable to differences in overall energy intake. We do not, however, discuss the estimated magnitude or level of intake for a subgroup based on the energy-adjusted results, since the within-subgroup variation in energy intake will affect the mean and median estimates.

Food Frequency Questionnaire

For the added sugars by food source models only, we incorporated covariates from the FFQ as additional predictors. FFQ data can improve the power of models to detect relationships between dietary intake and other variables by accounting for a portion of the variation in intake in the recall data (Subar et al., 2006). Since we used amount-only specifications for the nutrient models, inclusion of FFQ data was not appropriate in those cases. However, since our added sugars by food source models estimated the probability of consumption as part of the two-part approach described above, the inclusion of FFQ covariates was expected to yield improvements in model power.

For each added sugars food category that we analyzed, we included FFQ covariates indicating frequency of consumption of any foods within that category which could potentially contribute to added sugars intake. For example, in the specifications for the usual intake of added sugars from milk and milk products, we included a FFQ covariate measuring frequency of consumption of yogurt, but not an FFQ covariate measuring frequency of consumption of cheese, since cheese consumption is unlikely to contribute to added sugars intake. In some cases, a single FFQ covariate was included in multiple food source models. For example, the FFQ covariate measuring frequency of consumption of frozen yogurt, sorbets, or ices was included in both the milk and milk products and the sugar and sweets specifications, since it potentially applied to foods included in both of these categories.

Estimation Procedures

In both the amount-only and the two-part models, the amount data were first transformed to approximate normality using a nonlinear transformation for each individual in the sample. We then fit the model to obtain parameter estimates. A linear predictor of the amount of consumption was then estimated using the transformed data via a generalized linear model, with model covariates as described above. Parameter estimates from this model were then used as starting values in a nonlinear mixed model with person-specific random effects.

In the two-part models, the probability of consumption was additionally estimated via logistic regression, with the same set of model covariates as in the amount specification. Parameter estimates from the logistic regression were then used as starting values in a nonlinear mixed model with person-specific random effects. The probability model was linked with the amount model by using the parameter estimates from the two uncorrelated specifications as starting values for a model in which the two person-specific random effects were permitted to be correlated. These steps were carried out in SAS using the MIXTRAN macro developed by NCI staff. Next, Monte Carlo simulation (Metropolis and Ulam, 1949) was used to generate random effects for 100 pseudo-persons for each individual in the original sample. The random effect was then added to the linear predictor for each pseudo-person, and the amount estimates were back-transformed to the original scale with Taylor linearization. Means and percentiles were then estimated empirically from the resulting distribution. These steps were carried out in SAS using the NCI-supplied DISTRIB macro. Finally, standard errors of means and medians were obtained via Fay's balanced repeated replication (BRR) variant with a constant of 0.7 (Judkins, 1990). All BRR weights were post-stratified by age, sex, and race/ethnicity to match breakpoints aligning with the NHANES oversampling strategy.

Comparison with the Dietary Reference Intakes (DRIs)

As recommended for dietary assessment of groups, we generated the median and standard error of the median of the usual intakes of selected nutrients (IOM, 2000a) to enable the discussion of the nutrient intake of the added sugars/life stage subgroups in comparison to the appropriate DRI levels. The DRIs have specifically defined roles in dietary assessment, and their application varies with the nutrient under study based upon the DRI values available for that nutrient. For groups, the Estimated Average Requirement (EAR), which indicates the intake level expected to satisfy the needs of 50% of individuals in the group, is the recommended DRI as a cut-point for assessing potential nutrient inadequacy. For nutrients for which an EAR existed, we therefore estimated the percentage of each DRI life stage group with the usual intake less than the EAR for that group. For those nutrients for which no EAR existed, we estimated the percentage of each group with usual intake at or above the Adequate Intake (AI) level. ⁷ In general, the potential prevalence of inadequate intakes is likely to be low in groups with mean intake at or above the AI level. However, according to standard guidance no conclusions can be reached about nutrient inadequacy among individuals with mean intakes below the AI (IOM, 2000a).

Special care must be taken in interpreting estimated percentages relative to DRI levels in our models. The non-energy-adjusted results may be interpreted as estimates of the percentage of each population group of interest relative to the relevant DRI cut-point. However, one cannot determine from the non-energy-adjusted results to what extent differences in these estimates across groups are attributable to differences in total energy intake. If the total energy intake was identical across these groups, then the energy-adjusted estimated percentages relative to DRI cut-points would be identical to the non-energy-adjusted estimates. To the extent that the energy-adjusted results do differ from the non-energy-adjusted results, one can conclude that differences in the total energy intake are driving some portion of the observed variation across subgroups in the non-energy-adjusted estimates. However, the energy-adjusted percentage estimates should not be interpreted as estimates of the percentage of each life-stage group relative to the appropriate DRI cutoff because the amount of variation attributable to the total energy intake alone cannot be estimated.

RESULTS

Nutrient Analysis

Sample Distribution in Added Sugars Categories

We estimated that over 87 percent of the U.S. population had intakes of added sugars >0 ⩽25% of total energy intake (Table 1). In the >25% ⩽30% added sugars category the proportion of the population was 6.6%, with approximately 3% in each of the remaining added sugars categories, >30 ⩽35% and >35%, respectively, for a total of only approximately 13% of the population with added sugars intakes >25% of total energy. The majority of individuals had an estimated intake >5 ⩽20% of energy from added sugars, with the mean daily total energy intake from 2,063 kcal to 2,138 kcal. Controlling for the total energy intake in the analysis, this represented an estimated range of 45 to 92 mean gram-equivalents (g-eq) of added sugars intake daily. Males and females were evenly distributed in the total sample and did not differ appreciably in their distribution across the eight added sugars categories.

For each of the life stage groups, the largest proportion ingested >5 ⩽20% of energy as added sugars. More teenagers than any other life stage group consumed >15 ⩽20% of energy from added sugars (males and females 9 to 13 years (y): 31.2% and 27.8%, respectively; 14 to 18 y: 27.8% and 25.7%, respectively). A substantial number of males (22.7%) and females (17.5%) 14 to 18 y consumed >20 ⩽ 25% of their energy from added sugars. More females 14 to 18 y (20.3%) and 19 to 30 y (19.3%), followed by males 19 to 30 y (18.1%) and 14 to 18 y (17.0%) consumed >25% of their energy from added sugars than any other life stage group. Over two-thirds of older adults, 51+ y (68.4% to 73%) consumed ⩽15% of their energy from added sugars.

Compared to the overall population (12.5%), relatively more non-Hispanic Blacks (15.1%), persons with a Poverty-Income Ratio (PIR) below the poverty line (17.5%), and persons who were underweight (15.4%) consumed >25% of their energy from added sugars. Based on these NHANES data, overall an estimated 41% were normal or underweight based on body mass index (BMI: kg/m²: 39.3% and 1.6%, respectively) with approximately 30% in each of the overweight or obese categories. The individuals with the highest mean BMI values were associated with the ⩽ 0 ⩽ 5% and >35% added sugars categories (BMI: 28.9, 28.1, respectively). The highest percentage of persons who were overweight or obese consumed >5 ⩽ 15% of their energy from added sugars (>5 ⩽10% energy from added sugar: 22.5% overweight, 22.9% obese; >10 ⩽15% energy from added sugars: 24.3%overweight, 22.1% obese). With each 5% increase in added sugars category above 15% added sugars intake, we found a lower prevalence of overweight and obese individuals, with the exception of >35% added sugars for BMI⩾30 where the prevalence increased to 3.2%.

Estimated Usual Intake of Nutrients

We show weighted usual intake estimates, not controlling for total energy intake, for 13 nutrients plus total fiber for the 13 DRI life stage groups (Table 2). We also included the percent of the life stage group that is below the EAR or at/above the AI for each nutrient. Appendix Table 3 includes the same estimates with total energy intake as a covariate in the analysis. The nutrient intake profile is not strong for the United States population over four years, based on these 2003–2006 NHANES data in Table 2. Compared to the appropriate life stage EARs, the estimated median intakes for vitamin A, vitamin E, folate, and magnesium included the highest percentages of people at potential risk of nutrient inadequacy. Among nutrients for which there is only an AI, potassium and fiber had the lowest percent of the population across life-stage groups that exceeded the AI.

Table 2 Estimated Usual Intakes of Nutrients from Food [Median (SE)] by Dietary Reference Intake (DRI) Life Stage Groups ^a not including Total Energy as a Covariate for Vitamin A, Vitamin E, Vitamin C, Folate, Iron, Zinc, Magnesium, Phosphorus, Calcium, Sodium, Potassium, Vitamin K, Fiber and Total Choline ^b , with Percent below the DRI Estimated Average Requirement (EAR) or above the DRI Adequate Intake (AI), National Health and Nutrition Examination Survey (NHANES) 2003–2006.

	DRI Life Stage Groups
	Children	Males	Females	Males	Females	Males	Females	Males	Females	Males	Females	Males	Females
Nutrient	4 to 8 yrs	9 to 13 yrs	9 to 13 yrs	14 to 18 yrs	14 to 18 yrs	19 to 30 yrs	19 to 30 yrs	31 to 50 yrs	31 to 50 yrs	51 to 70 yrs	51 to 70 yrs	71+ yrs	71+ yrs
n (Total Choline n)^b	1,680 (808)	1,009 (484)	1,039(497)	1,351 (578)	1,250 (601)	1,097(455)	914 (419)	1,439 (672)	1,350 (643)	1,215 (559)	1,251 (579)	808 (334)	787 (296)
Median (SE)
Vitamin A, RAE	521 (18.9)	568 (15.0)	493 (19.0)	575 (15.3)	500 (20.0)	588 (13.1)	510 (19.1)	599 (11.8)	516 (18.1)	607 (12.2)	527 (17.2)	617 (15.0)	536 (17.4)
Percent<EAR (μg/d)	11%	30%	37%	58%	47%	55%	48%	54%	47%	53%	46%	51%	44%
Vitamin E, mg α-tocopherol	5.8 (0.08)	6.7 (0.09)	5.4 (0.07)	6.9 (0.08)	5.5 (0.08)	7.1 (0.08)	5.8 (0.07)	7.4 (0.09)	5.9 (0.08)	7.6 (0.11)	6.2 (0.09)	7.9 (0.11)	6.4 (0.10)
Percent<EAR (mg/d)	54%	80%	92%	94%	99%	93%	98%	92%	98%	91%	97%	88%	96%
Vitamin C, mg	70 (1.6)	75 (1.7)	67(1.5)	76 (1.4)	68(1.2)	78 (1.6)	70 (1.3)	80 (1.9)	71 (1.5)	81 (2.1)	72 (1.7)	82 (2.3)	74 (2.1)
Percent<EAR [mg/d]	5%	15%	20%	38%	38%	47%	40%	46%	39%	45%	38%	44%	37%
Folate, μg^c	389 (10.8)	445 (12.2)	344 (9.0)	445 (10.7)	343 (8.7)	443 (10.5)	345 (8.5)	446 (11.0)	348 (9.1)	446 (11.2)	349 (9.2)	449 (11.3)	350 (9.1)
Percent<EAR [μg/d]	1%	5%	19%	19%	46%	17%	42%	17%	41%	17%	41%	16%	40%
Iron, mg	15.2 (0.22)	17.6 (0.22)	13.1 (0.20)	17.5 (0.22)	13.1 (0.20)	17.6 (0.20)	13.1 (0.21)	17.6 (0.20)	13.0 (0.22)	17.5 (0.23)	13.0 (0.23)	17.5 (0.31)	13.0 (0.26)
Percent<EAR (mg/d)	0%	0%	1%	1%	7%	0%	8%	0%	8%	0%	0%	0%	0%
Zinc, mg	11.5 (0.20)	13.6 (0.25)	9.7 (0.18)	13.6 (0.24)	9.7 (0.18)	13.6 (0.21)	9.7 (0.17)	13.7 (0.19)	9.7 (0.17)	13.7 (0.19)	9.7 (0.16)	13.7 (0.22)	9.7 (0.18)
Percent<EAR (mg/d)	0%	2%	16%	7%	19%	12%	14%	12%	14%	11%	13%	11%	13%
Magnesium, mg	230 (3.1)	272 (3.6)	210 (2.7)	282 (3.6)	219 (3.0)	295 (3.3)	229 (3.0)	307 (3.1)	237 (3.1)	319 (3.3)	248 (3.1)	331 (3.9)	258 (3.5)
Percent<EAR (mg/d)	3%	17%	44%	72%	84%	64%	63%	66%	63%	62%	58%	57%	53%
Phosphorus, mg	1,305 (23.9)	1,500 (25.1)	1,117 (20.0)	1,490 (23.0)	1,109 (19.4)	1,489 (17.5)	1,106 (15.6)	1,485 (14.2)	1,097 (12.8)	1,477 (14.3)	1,095 (10.7)	1,471 (15.3)	1,091 (11.6)
Percent<EAR [mg/d]	0%	12%	43%	12%	44%	0%	3%	0%	3%	0%	3%	0%	4%
Calcium, mg	1,012 (21.3)	1,083 (22.1)	873 (16.1)	1,039 (20.7)	836 (15.7)	1,004 (16.5)	804 (13.4)	968 (13.0)	768 (11.8)	928 (12.2)	740 (10.7)	892 (13.8)	709 (10.7)
Percent⩾AI (mg/d)	72%	30%	14%	27%	11%	50%	29%	47%	25%	24%	10%	21%	8%
Sodium, mg	3,354 (42.1)	3,856 (59.9)	2,885 (29.4)	3,837 (59.1)	2,871 (29.9)	3,843 (49.2)	2,868 (24.6)	3,839 (40.9)	2,852 (21.0)	3,824 (38.9)	2,852 (19.7)	3,815 (42.4)	2,845 (21.8)
Percent⩾AI [mg/d]	100%	100%	97%	100%	97%	100%	97%	100%	97%	100%	99%	100%	99%
Potassium, mg	2,192 (35.3)	2,594 (40.6)	2,003 (32.4)	2,691 (42.3)	2,087 (36.2)	2,808 (37.8)	2,182 (35.2)	2,922 (35.0)	2,268 (34.2)	3,029 (35.9)	2,369 (33.2)	3,144 (42.3)	2,468 (35.7)
Percent⩾AI [mg/d]	3%	2%	0%	2%	0%	2%	0%	3%	0%	4%	0%	6%	1%
Vitamin K, μg	49 (1.1)	57 (1.4)	52 (1.2)	63 (1.7)	58 (1.3)	71 (2.3)	65 (1.5)	79 (3.1)	72 (2.0)	88 (3.9)	80 (2.6)	98 (4.5)	89 (3.3)
Percent⩾AI [μg/d]	41%	46%	39%	37%	30%	14%	26%	20%	32%	26%	41%	34%	49%
Total Fiber, g	13 (0.3)	15(0.4)	12(0.3)	15(0.4)	12 (0.3)	16 (0.4)	13 (0.3)	16 (0.4)	14 (0.3)	17 (0.4)	14 (0.4)	18 (0.5)	15 (0.4)
Percent⩾AI [g/d]	3%	1%	1%	0%	1%	0%	3%	0%	3%	4%	13%	5%	15%
Total Choline, mg^b	252 (5.6)	320 (6.1)	221 (3.8)	337 (6.5)	233 (3.4)	350 (5.7)	245 (3.9)	368 (6.0)	259 (3.6)	385 (5.4)	272 (4.1)	404 (7.3)	287 (4.6)
^aBased on the Dietary Reference Intake Life Stage Groups used in the Dietary Reference Intakes (Institute of Medicine (IOM), 1997) excluding children under 4 years, fasters, and pregnant and lactating women. Values were NOT controlled for total energy intake during this analyses.
^bTotal choline values available for NHANES 2005–2006 only and sample sizes for choline are shown in parentheses. Total choline content was calculated as the sum of free choline (Cho); glycerophosphocholine (GPC); phosphocholine (Pcho); phosphatidylcholine (PTdcho); and sphingomyelin (SM) (Patterson et al, 2008). Sample sizes for total choline were not adequate to generate percent greater than the AI.
^cThe NHANES provides values for folic acid and food folate in Dietary Folate Equivalents (DFE). We present folate as Dietary Folate Equivalents. IDFE = 1 μg food folate = 0.6 mg of folic acid from fortified food.

For some nutrients, a higher risk of nutrient inadequacy was specific to certain life stages. For phosphorus, while only 0 to 4% of each adult life stage group had estimated intakes below the EAR, for children ages 9 to18 y, 12 to 44% had estimated intakes below the EAR. Compared to all other life-stage groups, females ages 9 to 18 y had the highest proportion of intake below the EAR for zinc (16 and 19%). By contrast, with the exception of children 4 to 8 y, all other life stage groups had a considerable number of individuals with intakes below their EARs for magnesium (17% to 84%) and vitamin C (15 to 46%). Females 14 to 71+ y had lower intakes (343 to 446 ug/d) and a higher percent of each life stage group below the EAR (40–46%) for folate than males (443 to 449 ug/d; 16 to17%). This gender difference was retained when the total energy intake was included as a covariate in the analyses, indicating that the gender difference was attributable to differences in some additional aspects of dietary composition rather than differences in total amount of calories consumed (Appendix Table 3). Iron is the only nutrient for which <10% of all life-stage groups had intakes below the EAR.

Among nutrients with an AI, potassium was the nutrient with the overall highest percentages of people at potential risk of nutrient inadequacy, with only 0% to 6% of intakes across the life stage groups exceeding the AI. Only 8% to 50% of individuals in each life-stage group, except for children 4 to 8 y (72%), had calcium intakes above their respective AI. The proportion is especially low for females 9 to13 y (14%), 14 to 18 y (11%), 51 to 70 y (10%), and 71+ y (8%). Depending on the life-stage group, 14% to 49% of individuals had vitamin K intakes meeting or exceeding the respective AI. In contrast, for all life stage groups, ⩾97% of individuals met or exceeded the AI for sodium, and the median intakes for all adult life-stage groups exceeded their respective Tolerable Upper Intake Level (UL).

The estimated total fiber intakes were very low. With the exception of older women (51+ y), only 0 to 5% of individuals in all other life stage groups had fiber intakes meeting or exceeding the AI. The estimated total fiber intake in grams was lower in females (12 to 15 g/d) than males (15 to 18 g/d) for all life-stage groups. Controlling for total energy intake reduced the gender difference (Appendix Table 3), indicating that differences in total calories consumed by each gender account for a portion of this fiber intake gap.

We were not able to generate estimates of the percent of individuals with intakes greater than the AI for total choline because the sample sizes were too small. However, median choline intakes for all life-stage groups, with the exception of children 4 to 8 y, were appreciably lower than the respective AIs.

Including the total energy intake as a covariate and comparing median nutrient intake by sequential two year data collection periods of the NHANES (2001–2002; 2003–2004; 2005–2006), we found that intakes were higher with each successive time period overall except for vitamin C, phosphorus, potassium, and total fiber. ⁸ This indicates that the nutrient profile for these nationally weighted intake estimates appears to have improved somewhat over the 2001–2006 interval, and that any estimated increase in nutrient intake was not purely attributable to increases in total calorie intake over the same period.

Estimated Usual Intakes by Percent Daily Intake of Added Sugars: Age Groups Combined

For all DRI age groups combined, beginning with the >5 ⩽10% added sugars category, for all nutrients, including fiber, the median nutrient intake was less with each 5%-interval increase in added sugars intake (Table 3). The median intake was over 30 percent lower in the highest added sugars category as compared to the lowest added sugars category for the following nutrients: vitamin A (45%), vitamin E (45%), vitamin C (47%), folate (41%), zinc (34%), magnesium (48%), phosphorus (36%), calcium (31%), potassium (46%), vitamin K (60%), fiber (50%), and total choline (63%). Appendix Table 4 includes estimates for the same nutrients and age groups without total energy as a covariate in the model.

Table 3 Estimated Usual Intakes of Nutrients from Food [Median (SE)] by Range of Percent of Estimated Daily Intake from Added Sugars Including Total Energy Intake as a Covariate for all Life Stages Combined ^a for Vitamin A, Vitamin E, Vitamin C, Folate, Iron, Zinc, Magnesium, Phosphorus, Calcium, Sodium, Potassium, Vitamin K, Fiber and Total Choline ^b , National Health and Nutrition Examination Survey (NHANES) 2003–2006

	Categories of Individuals Based on Percent of Energy Intake from Added Sugar^c
Nutrient	0 ⩽ x ⩽ 5%	5 < x ⩽ 10%	10 < x ⩽ 15%	15 < x ⩽ 20%	20 < x ⩽ 25%	25 < x ⩽ 30%	30 < x ⩽ 35%	>35%
n (Total Choline n)^b	1,387 (645)	2,877 (1,372)	3,666 (1,725)	3,311 (1,523)	2,055 (909)	1,078 (458)	432 (164)	384 (129)
Median (SE)
Vitamin A, μg RAE	603 (13.8)	588 (14.3)	552 (9.6)	510 (22.3)	482 (17.1)	449 (15.1)	393 (15.7)	330 (13.3)
Vitamin E, mg α-tocopherol	6.7 (0.15)	6.6 (0.09)	6.2 (0.10)	5.8 (0.20)	5.5 (0.12)	5.2 (0.11)	4.5 (0.16)	3.7 (0.13)
Vitamin C, mg	85 (2.6)	81 (1.9)	76 (1.8)	69 (1.5)	65 (1.4)	60 (1.8)	53 (1.7)	45 (2.4)
Folate, μg^d	398 (5.5)	394 (6.3)	373 (3.4)	350 (12.8)	334 (9.2)	314 (8.1)	276 (7.5)	234 (7.2)
Iron, mg	14.6 (0.25)	14.8 (0.21)	14.5 (0.15)	13.9 (0.52)	13.7 (0.37)	13.2 (0.34)	12.0 (0.32)	10.5 (0.30)
Zinc, mg	11.2 (0.21)	11.3 (0.17)	10.9 (0.14)	10.4 (0.43)	10.2 (0.29)	9.6 (0.29)	8.7 (0.24)	7.4 (0.29)
Magnesium, mg	285 (4.1)	278 (3.4)	260 (3.2)	240 (8.1)	227 (5.3)	209 (4.6)	181 (5.2)	148 (5.0)
Phosphorus, mg	1,267 (18.7)	1,279 (22.0)	1,236 (18.8)	1,177 (42.8)	1,158 (24.3)	1,094 (22.0)	979 (38.6)	813 (44.5)
Calcium, mg	865 (13.5)	879 (16.0)	852 (9.9)	816 (27.1)	799 (16.1)	769 (16.0)	690 (25.5)	593 (30.7)
Sodium, mg	3,184 (50.9)	3,251 (53.0)	3,177 (50.8)	3,063 (101.7)	3,048 (54.3)	2,921 (51.8)	2,661 (111.1)	2,268 (124.4)
Potassium, mg	2,710 (33.9)	2,656 (35.0)	2,497 (25.2)	2,321 (86.2)	2,203 (57.5)	2,037 (51.7)	1,774 (53.6)	1,451 (44.6)
Vitamin K, μg	86 (2.5)	81 (1.9)	72 (2.3)	63 (1.3)	57 (1.6)	51 (2.0)	42 (2.2)	34 (2.4)
Fiber, g	16 (0.2)	15 (0.3)	14 (0.2)	13 (0.5)	12 (0.3)	11 (0.3)	10 (0.3)	8 (0.3)
Total Choline, mg^b	319 (6.1)	313 (9.4)	285 (6.2)	273 (5.7)	254 (10.6)	242 (9.3)	207 (31.1)	177 (19.2)
^aBased on the Dietary Reference Intake Life Stage Groups (Institute of Medicine (IOM) 1997) excluding children under 4 years, fasters, and pregnant and lactating women. These analyses controlled for total energy intake.
^bTotal choline values available for NHANES 2005–2006 only and sample sizes for choline are shown in parentheses. Total choline content was calculated as the sum of free choline (Cho); glycerophosphocholine (GPC); phosphocholine (Pcho); phosphatidylcholine (PTdcho); and sphingomyelin (SM) (Patterson et al., 2008). Sample sizes for total choline were not adequate to generate percent greater than the AI.
^cOur analyses employed two days of 24-hour dietary recall data from the National Health and Nutrition Examination Survey (NHANES), What We Eat in America 2003–2006. We restricted our sample to children and adults as defined by the DRIs as aged 4 years or older. Our sample was further restricted to individuals with reliable recall status, excluding fasters, pregnant women, and lactating women. The final analytic sample included 29,099 days of recall data from 15,189 individual respondents. For 2,539 individuals in the sample, only one 24-hour recall was reported.
^dThe NHANES provides values for folic acid and food folate in Dietary Folate Equivalents (DFE). We present folate as Dietary Folate Equivalents. IDFE = 1 μg food folate = 0.6 mg of folic acid from fortified food.

Estimated Usual Intakes by Percent Daily Intake of Added Sugars: DRI Age Groups

Intakes of nutrients from food by three added sugars categories are shown for children 4 y and older ⁹ (Table 4) and adults (Table 5) by DRI life-stage group with the appropriate DRI recommended intake levels for each nutrient (IOM, 2000a). The total energy intake was not included as a covariate for the usual intake estimates in these two tables; the estimates adjusted for the total energy intake can be found in Appendix Tables 5 and 6.

Table 4 Estimated Usual Intakes of Nutrients from Food [Median (SE)] by Range of Percent of Estimated Daily Intake from Added Sugars without Total Energy Intake as a Covariate in the Analysis, by Child Dietary Reference Intake (DRI) Life Stage Groups ^a for Vitamin A, Vitamin E, Vitamin C, Folate, Iron, Zinc, Magnesium, Phosphorus, Calcium, Sodium, Potassium, Vitamin K, Fiber and Total Choline ^b , National Health and Nutrition Examination Survey (NHANES) 2003–2006

	Categories of Individuals Based on Percent of Energy Intake from Added Sugar^c
	Children 4 to 8 Yrs			Males 9 to 13 Yrs			Females 9 to 13 Yrs			Males 14 to 18 Yrs			Females 14 to 18 Yrs
Nutrient	0 ⩽ x ⩽ 15%	15 < x ⩽ 25%	>25%	0 ⩽ x ⩽ 15%	15 < x⩽ 25%	>25%	0 ⩽ x ⩽ 15%	15 < x⩽ 25%	>25%	0 ⩽ x ⩽ 15%	15 < x ⩽ 25%	>25%	0 ⩽ x ⩽ 15%	15 < x ⩽ 25%	>25%
N(Total Choline n)^b	840 (421)	694 (333)	146 (54)	433 (212)	450 (228)	126 (44)	477 (250)	445 (203)	117 (44)	470 (222)	620 (266)	261 (90)	449 (239)	568 (257)	233 (105)
Median (SE)
Vitamin A, μg RAE	597 (20.4)	507 (18.5)	427 (20.0)	650 (13.3)	543 (15.9)	455 (20.9)	557 (19.7)	470 (18.4)	388 (21.7)	647 (11.4)	545 (13.1)	455 (15.3)	559 (22.4)	472 (16.5)	393 (22.8)
Percent<EAR	(275 μg/d)^d 6%	11%	20%	(445 μg/d) 20%	34%	48%	(420 μg/d) 27%	41%	55%	(630 μg/d) 48%	63%	76%	(485 μg/d) 38%	52%	67%
Vitamin E, mg α-tocopherol	6.5 (0.10)	5.7 (0.09)	4.9 (0.11)	7.5 (0.14)	6.4 (0.08)	5.5 (0.14)	6.0 (0.09)	5.2 (0.07)	4.4 (0.12)	7.6 (0.14)	6.6 (0.09)	5.7 (0.11)	6.1 (0.08)	5.3 (0.07)	4.6 (0.12)
Percent<EAR	(6 mg/d) 42%	56%	70%	(9 mg/d) 71%	83%	92%	(9 mg/d) 88%	94%	97%	(12 mg/d) 91%	96%	98%	(12 mg/d) 97%	99%	100%
Vitamin C, mg	82 (2.5)	68 (1.5)	56 (1.7)	88 (3.5)	71 (1.4)	58 (2.2)	77 (2.2)	64 (1.4)	51 (2.2)	87 (3.4)	72 (1.6)	58 (2.0)	77 (1.7)	64 (1.4)	52 (2.2)
Percent<EAR	(22 mg/d) 2%	5%	9%	(39 mg/d) 9%	17%	27%	(39 mg/d) 13%	22%	33%	(63 mg/d) 29%	41%	55%	(56 mg/d) 29%	41%	55%
Folate, μg^e	429 (7.6)	374 (5.9)	323 (8.4)	486 (6.3)	417 (6.7)	358 (10.7)	374 (6.0)	322 (5.5)	273 (7.5)	479 (5.1)	413 (5.7)	354 (8.0)	371 (7.1)	319 (4.9)	272 (8.8)
Percent<EAR	(160 μg/d) 0%	1%	4%	(250 μg/d) 2%	6%	15%	(300 μg/d) 11%	23%	40%	(330 μg/d) 12%	24%	42%	(400 μg/d) 36%	54%	72%
Iron, mg	16.3 (0.33)	15.1 (0.20)	13.9 (0.40)	18.9 (0.29)	17.2 (0.25)	15.7 (0.48)	13.9 (0.20)	12.8 (0.20)	11.6 (0.32)	18.6 (0.23)	17.0 (0.22)	15.6 (0.37)	13.8 (0.21)	12.7 (0.17)	11.6 (0.38)
Percent<EAR	(4.1 mg/d) 0%	0%	0%	(5.9 mg/d) 0%	0%	0%	(5.7 mg/d) 0%	1%	2%	(7.7 mg/d) 0%	1%	1%	(7.9 mg/d) 4%	8%	13%
Zinc, mg	12.5 (0.25)	11.3 (0.24)	10.2 (0.38)	14.9 (0.22)	13.2 (0.28)	11.8 (0.43)	10.5 (0.16)	9.4 (0.18)	8.3 (0.27)	14.7 (0.17)	13.1 (0.24)	11.7 (0.34)	10.4 (0.18)	9.3 (0.15)	8.3 (0.30)
Percent<EAR	(4.0 mg/d) 0%	0%	0%	(7.0 mg/d) 1%	2%	5%	(7.0 mg/d) 9%	18%	30%	(8.5 mg/d) 3%	8%	15%	(7.3 mg/d) 13%	23%	35%
Magnesium, mg	262 (3.9)	225 (3.2)	191 (5.4)	309 (3.4)	261 (4.0)	221 (6.1)	237 (2.4)	201 (2.6)	167 (3.9)	316 (3.2)	269 (3.4)	227 (4.8)	244 (2.9)	207 (2.2)	174 (4.4)
Percent<EAR	(110 mg/d) 1%	2%	7%	(200 mg/d) 7%	19%	38%	(200 mg/d) 29%	50%	71%	(340 mg/d) 60%	79%	91%	(300 mg/d) 75%	90%	96%
Phosphorus, mg	1,419 (32.7)	1,289 (26.0)	1,159 (33.3)	1,632 (26.4)	1,459 (25.7)	1,308 (27.4)	1,211 (26.0)	1,081 (17.6)	952 (17.7)	1,605 (20.0)	1,442 (18.2)	1,288 (18.6)	1,195 (26.7)	1,067 (12.0)	945 (17.5)
Percent<EAR	(405 mg/d) 0%	0%	0%	(1,055 mg/d) 6%	13%	24%	(1,055 mg/d) 32%	47%	62%	(1,055 mg/d) 7%	14%	25%	(1,055 mg/d) 33%	49%	64%
Calcium, mg	1,108 (32.2)	997 (20.3)	892 (18.0)	1,189 (26.1)	1,051 (21.2)	933 (19.9)	949 (23.8)	844 (13.5)	739 (13.9)	1,128 (20.3)	1,002 (15.3)	886 (15.5)	904 (25.2)	803 (9.5)	708 (12.8)
Percent⩾AI	(800 mg/d) 80%	71%	61%	(1300 mg/d) 40%	27%	17%	(1,300 mg/d) 19%	11%	6%	(1,300 mg/d) 34%	23%	15%	(1,300 mg/d) 16%	9%	5%
Sodium, g	3,545 (60.8)	3,335 (44.8)	3,113 (78.9)	4,093 (61.5)	3,783 (62.5)	3,513 (76.0)	3,045 (41.5)	2,823 (27.1)	2,589 (37.8)	4,041 (49.2)	3,754 (49.6)	3,476 (53.7)	3,018 (48.8)	2,799 (19.0)	2,583 (35.1)
Percent⩾AI	(1.2 g/d) 100%	100%	99%	(1.5 g/d) 100%	100%	100%	(1.5 g/d) 98%	97%	95%	(1.5 g/d) 100%	100%	100%	(1.5 g/d) 98%	97%	94%
Potassium, g	2,481 (35.7)	2,145 (40.2)	1,825 (70.9)	2,918 (31.7)	2,496 (47.9)	2,129 (76.0)	2,248 (25.1)	1,913 (34.7)	1,596 (55.4)	2,983 (28.3)	2,569 (41.2)	2,189 (61.3)	2,315 (30.7)	1,974 (31.1)	1,662 (62.0)
Percent⩾AI	(3.8 g/d) 6%	2%	1%	(4.5 g/d) 4%	1%	0%	(4.5 g/d) 0%	0%	0%	(4.7 g/d) 3%	1%	0%	(4.7 g/d) 0%	0%	0%
Vitamin K, μg	59 (1.5)	47 (1.0)	38 (1.0)	69 (1.7)	54 (1.3)	43 (1.3)	62 (1.5)	49 (1.1)	39 (1.2)	75 (2.2)	59 (1.9)	47 (2.0)	67 (1.6)	54 (1.3)	43 (1.0)
Percent⩾AI	(55 μg/d) 56%	38%	21%	(60 μg/d) 61%	41%	24%	(60 μg/d) 52%	33%	18%	(75 μg/d) 49%	31%	16%	(75 μg/d) 41%	24%	12%
Total Fiber, g	14 (0.2)	12 (0.2)	10 (0.2)	16 (0.3)	14 (0.2)	12 (0.3)	13 (0.2)	11 (0.2)	9 (0.2)	17 (0.3)	14 (0.2)	12 (0.3)	14 (0.2)	11 (0.2)	10 (0.3)
Percent⩾AI (g/d)	(25 g/d) 4%	2%	0%	(31 g/d) 2%	0%	0%	(26 g/d) 2%	0%	0%	(38 g/d) 0%	0%	0%	(26 g/d) 2%	1%	0%
Total Choline, mg^b	282 (9.1)	240 (7.0)	207 (22.8)	353 (9.9)	306 (6.7)	266 (9.8)	244 (7.0)	208 (3.6)	176 (7.3)	370 (10.6)	318 (6.5)	274 (10.3)	253 (7.3)	217 (3.7)	185 (10.6)
^aBased on the Dietary Reference Intake Life Stage Groups used in the Dietary Reference Intakes (Institute of Medicine (IOM), 1997) excluding children under 4 years, fasters, and pregnant and lactating women. Values were controlled for total energy intake during the analyses.
^bTotal choline values available for NHANES 2005–2006 only and sample sizes for choline are shown in parentheses. Total choline content was calculated as the sum of free choline (Cho); lycerophosphocholine (GPC); phosphocholine (Pcho); phosphatidylcholine (PTdcho); and sphingomyelin (SM) (Patterson et al., 2008). Sample sizes for total choline were not adequate to generate percent greater than the AI.
^cOur analyses employed two days of 24-hour dietary recall data from the National Health and Nutrition Examination Survey (NHANES), What We Eat in America 2003–2006. We restricted our sample to children and adults as defined by the DRIs as aged 4 years or older. Our sample was further restricted to individuals with reliable recall status, excluding fasters, pregnant women, and lactating women. The final analytic sample included 29,099 days of recall data from 15,189 individual respondents. For 2,539 individuals in the sample, only one 24-hour recall was reported.
^dEAR or AI for the respective life stage group.
^eThe NHANES provides values for folic acid and food folate in Dietary Folate Equivalents (DFE). We present folate as Dietary Folate Equivalents. IDFE = 1 μg food folate = 0.6 mg of folic acid from fortified food.

Among children less than 18 y (Table 4), and adults of all ages (Table 5), for all nutrients, the median estimated nutrient intakes were lower with increased added sugars intake and the nutrient intake was lowest with added sugars intakes greater than 25%.

Table 5 Estimated Usual Intakes of Nutrients from Food [Median (SE)] by Range of Percent of Estimated Daily Intake from Added Sugars without Total energy Intake as a Covariate in the Analysis, by Adult Dietary Reference Intake (DRI) Life Stage Groups ^a for Vitamin A, Vitamin E, Vitamin C, Folate, Iron, Zinc, Magnesium, Phosphorus, Calcium, Sodium, Potassium, Vitamin K, Fiber and Total Choline ^b , National Health and Nutrition Examination Survey (NHANES) 2003–2006

	Categories of Individuals Based on Percent of Energy Intake from Added Sugar^c
	Males 19 to 30 Yrs			Females 19 to 30 Yrs			Males 31 to 50 Yrs			Females 31 to 50 Yrs			Males 51 to 70 Yrs			Females 51 to 70 Yrs			Males 71+ Yrs			Females 71+ Yrs
Nutrient	0 ⩽ x⩽ 15%	15 < x⩽25%	>25%	0 ⩽ x⩽15%	15 < x⩽ 25%	>25%	0 ⩽ x⩽ 15%	15 < x⩽ 25%	>25%	0 ⩽ x⩽ 15%	15 < x⩽ 25%	>25%	0 ⩽ x⩽ 15%	15 < x⩽ 25%	>25%	0 ⩽ x⩽ 15%	15 < x⩽ 25%	>25%	0 ⩽ x⩽ 15%	15 < x⩽ 25%	>25%	0 ⩽ x⩽ 15%	15 < x⩽ 25%	>25%
n (Total Choline n)^b	495(201)	398(173)	204(81)	421(211)	323(143)	170(65)	807(377)	461(221)	171(74)	781(396)	383(177)	186(70)	817(368)	296(140)	102(51)	839(397)	298(130)	114(52)	562(238)	210(83)	36(13)	539(210)	220(78)	28(8)
Median (SE)
Vitamin A, μg RAE	650(12.6)	546(13.0)	461(12.9)	561(19.9)	470(20.7)	393(20.2)	652(10.9)	551(10.6)	463(14.1)	564(15.2)	475(18.9)	394(16.5)	660(12.0)	550(12.8)	464(14.9)	564(18.6)	473(16.2)	399(19.3)	657(16.4)	554(11.0)	471(10.7)	575(15.2)	482(16.0)	399(17.7)
Percent<EAR	625 μg/d^d 47%	62%	75%	500 μg/d 40%	55%	70%	625 μg/d 46%	61%	74%	500 μg/d 39%	54%	69%	625 μg/d 45%	61%	73%	500 μg/d 40%	55%	69%	625 μg/d 45%	61%	74%	500 μg/d 38%	53%	69%
Vitamin E, mg α-tocopherol	7.8(0.10)	6.7(0.09)	5.8(0.10)	6.2(0.08)	5.4(0.09)	4.6(0.11)	7.9(0.13)	6.9(0.10)	6.0(0.12)	6.4(0.07)	5.5(0.09)	4.7(0.10)	8.2(0.14)	7.0(0.10)	6.1(0.14)	6.5(0.09)	5.6(0.09)	4.9(0.13)	8.3(0.13)	7.2(0.13)	6.3(0.14)	6.7(0.10)	5.8(0.10)	5.0(0.13)
Percent<EAR	12 mg/d 90%	96%	98%	12 mg/d 97%	99%	100%	12 mg/d 89%	95%	98%	12 mg/d 97%	99%	100%	12 mg/d 88%	94%	97%	12 mg/d 96%	99%	100%	12 mg/d 86%	93%	97%	12 mg/d 96%	98%	100%
Vitamin C, mg	87(2.7)	72(1.6)	59(2.0)	78(1.6)	64(1.6)	52(2.2)	88(3.1)	72(1.8)	59(2.3)	78(2.0)	65(1.6)	52(2.1)	89(2.9)	72(1.9)	59(2.6)	78(1.8)	64(1.8)	53(2.7)	88(2.5)	73(2.5)	61(2.9)	80(2.3)	65(2.0)	53(2.7)
Percent<EAR	75 mg/d 39%	53%	66%	60 mg/d 33%	46%	59%	75 mg/d 39%	52%	65%	60 mg/d 32%	45%	59%	75 mg/d 38%	53%	65%	60 mg/d 33%	46%	58%	75 mg/d 39%	52%	64%	60 mg/d 31%	44%	58%
Folate, μg ^e	475(5.0)	408(6.1)	352(7.5)	367(6.4)	314(7.3)	268(8.1)	470(4.8)	406(5.6)	349(8.6)	364(4.9)	313(7.1)	265(6.8)	468(5.3)	401(7.3)	345(9.6)	360(7.0)	308(6.5)	264(8.7)	462(8.5)	398(6.0)	344(6.3)	360(6.3)	308(7.1)	260(8.4)
Percent<EAR	320 μg/d 11%	22%	39%	400 μg/d 34%	52%	71%	320 μg/d 11%	23%	40%	400 μg/d 34%	53%	71%	320 μg/d 12%	24%	41%	400 μg/d 36%	55%	73%	320 μg/d 12%	25%	41%	400 μg/d 36%	55%	73%
Iron, mg	18.5(0.17)	16.9(0.24)	15.5(0.36)	13.7(0.18)	12.5(0.28)	11.4(0.37)	18.3(0.17)	16.8(0.23)	15.4(0.42)	13.6(0.14)	12.5(0.28)	11.3(0.32)	18.3(0.18)	16.6(0.32)	15.3(0.47)	13.5(0.22)	12.3(0.27)	11.3(0.42)	18.1(0.30)	16.5(0.26)	15.3(0.30)	13.5(0.21)	12.3(0.30)	11.2(0.41)
Percent<EAR	6.0 mg/d 0%	0%	0%	8.1 mg/d 5%	9%	15%	6.0 mg/d 0%	0%	0%	8.1 mg/d 6%	9%	16%	6.0 mg/d 0%	0%	0%	5.0 mg/d 0%	0%	1%	6.0 mg/d 0%	0%	0%	5.0 mg/d 0%	0%	1%
Zinc, mg	14.6(0.16)	13.0(0.23)	11.7(0.31)	10.3(0.15)	9.2(0.21)	8.2(0.27)	14.5(0.15)	13.0(0.21)	11.6(0.33)	10.3(0.11)	9.2(0.20)	8.1(0.23)	14.4(0.17)	12.8(0.24)	11.5(0.34)	10.2(0.16)	9.1(0.18)	8.1(0.28)	14.3(0.21)	12.8(0.22)	11.5(0.26)	10.2(0.14)	9.1(0.20)	8.0(0.26)
Percent<EAR	9.4 mg/d 7%	14%	24%	6.8 mg/d 9%	17%	28%	9.4 mg/d 8%	15%	25%	6.8 mg/d 9%	17%	29%	9.4 mg/d 8%	15%	25%	6.8 mg/d 10%	18%	28%	9.4 mg/d 8%	15%	27%	6.8 mg/d 10%	17%	29%
Magnesium, mg	325(2.8)	276(3.4)	235(4.3)	251(2.4)	212(3.4)	178(4.1)	333(3.3)	285(3.2)	242(4.9)	259(2.2)	219(3.3)	184(3.5)	344(3.8)	292(3.5)	249(5.2)	265(2.6)	225(3.2)	190(4.6)	352(3.7)	300(4.2)	258(4.6)	275(3.2)	233(3.4)	196(4.4)
Percent<EAR	330 mg/d 52%	73%	87%	255 mg/d 52%	73%	88%	350 mg/d 57%	76%	90%	265 mg/d 53%	74%	88%	350 mg/d 52%	74%	87%	265 mg/d 50%	71%	86%	350 mg/d 49%	71%	87%	265 mg/d 45%	67%	83%
Phosphorus, mg	1,587(19.4)	1,423(15.7)	1,279(17.1)	1,180(20.3)	1,048(15.5)	929(12.9)	1,568(14.5)	1,410(13.9)	1,263(17.5)	1,166(11.9)	1,039(11.5)	914(13.0)	1,558(16.1)	1,388(14.0)	1,246(19.0)	1,149(13.6)	1,019(10.7)	907(13.8)	1,532(17.1)	1,374(23.2)	1,239(34.9)	1,144(12.2)	1,014(12.2)	891(17.1)
Percent<EAR	580 mg/d 0%	0%	1%	580 mg/d 2%	4%	9%	580 mg/d 0%	0%	1%	580 mg/d 2%	4%	10%	580 mg/d 0%	0%	1%	580 mg/d 2%	5%	10%	580 mg/d 0%	0%	1%	580 mg/d 2%	5%	10%
Calcium, mg	1,077(20.9)	954(12.9)	849(15.3)	861(20.3)	759(12.9)	669(10.3)	1,028(14.7)	913(9.9)	808(13.4)	820(12.8)	725(10.0)	633(12.9)	987(14.3)	865(10.2)	767(13.8)	778(15.2)	684(9.2)	605(12.5)	934(19.0)	826(13.3)	736(25.0)	748(11.4)	657(9.5)	571(14.4)
Percent⩾AI	1,000 mg/d 58%	45%	33%	1,000 mg/d 34%	23%	14%	1,000 mg/d 53%	40%	28%	1,000 mg/d 30%	20%	12%	1,200 mg/d 29%	18%	12%	1,200 mg/d 12%	7%	3%	1,200 mg/d 25%	15%	8%	1,200 mg/d 10%	5%	2%
Sodium, mg	4,012(52.4)	3,719(44.2)	3,467(50.8)	2,994(37.1)	2,763(24.0)	2,552(29.0)	3,981(40.9)	3,702(36.2)	3,440(48.3)	2,972(23.4)	2,752(17.8)	2,526(35.7)	3,971(43.3)	3,662(36.9)	3,409(49.2)	2,942(28.5)	2,714(22.3)	2,518(32.2)	3,921(53.1)	3,640(43.1)	3,405(73.9)	2,944(26.5)	2,715(25.5)	2,489(40.1)
Percent⩾AI	1.5 g/d 100%	100%	99%	1.5 g/d98%	96%	94%	1.5 g/d100%	100%	100%	1.5 g/d98%	96%	94%	1.3 g/d 100%	100%	100%	1.3 g/d99%	99%	97%	1.2 g/d 100%	100%	100%	1.2 g/d 100%	99%	98%
Potassium, mg	3,067(28.2)	2,639(42.9)	2,268(55.7)	2,385(26.0)	2,028(44.6)	1,712(57.3)	3,149(27.9)	2,723(37.8)	2,338(62.3)	2,459(19.1)	2,101(42.1)	1,767(48.4)	3,249(31.6)	2,792(43.2)	2,405(65.7)	2,526(27.5)	2,156(37.1)	1,836(59.4)	3,319(38.4)	2,875(38.2)	2,493(43.9)	2,622(26.6)	2,241(38.5)	1,891(56.1)
Percent⩾AI	4.7 g/d4%	1%	0%	4.7 g/d0%	0%	0%	4.7 g/d5%	1%	0%	4.7 g/d0%	0%	0%	4.7 g/d6%	1%	1%	4.7 g/d1%	0%	0%	4.7 g/d7%	2%	0%	4.7 g/d1%	0%	0%
Vitamin K, μg	82(2.4)	65(2.4)	52(2.6)	74(1.6)	59(1.2)	47(1.5)	89(3.6)	71(3.3)	57(3.0)	81(2.2)	65(1.8)	51(2.2)	98(4.4)	77(3.7)	62(3.7)	88(2.4)	70(2.7)	56(2.5)	107(4.3)	85(5.5)	68(5.9)	98(3.8)	78(3.4)	61.3(3.3)
Percent⩾AI	120 μg/d 20%	9%	4%	90 μg/d 33%	18%	8%	120 μg/d 27%	13%	5%	90 μg/d 41%	24%	12%	120 μg/d 33%	17%	9%	90 μg/d 48%	30%	15%	120 μg/d 40%	22%	10%	90 μg/d 57%	37%	21%
Total Fiber, g	17(0.2)	15(0.2)	12(0.2)	14(0.2)	12(0.3)	10(0.3)	18(0.2)	15(0.2)	13(0.3)	14(0.2)	12(0.3)	10(0.3)	18(0.2)	15(0.3)	13(0.3)	15(0.3)	12(0.3)	11(0.4)	19(0.4)	16(0.2)	14(0.2)	15(0.3)	13(0.4)	11(0.4)
Percent⩾AI	38 g/d0%	0%	0%	25 g/d3%	1%	0%	38 g/d0%	0%	0%	25 g/d4%	1%	0%	30 g/d4%	1%	1%	21 g/d14%	6%	2%	30 g/d5%	2%	0%	21 g/d16%	7%	2%
Total Choline, mg^b	382(7.5)	330(6.8)	285(12.8)	265(4.6)	227(5.4)	194(11.3)	395(7.2)	344(7.5)	297(15.8)	278(4.4)	236(6.4)	203(10.2)	410(5.9)	356(7.5)	311(21.7)	288(3.0)	248(6.7)	211(12.4)	425(5.1)	371(10.2)	331(25.3)	299(3.7)	257(9.5)	219(12.1)
^aBased on the Dietary Reference Intake Life Stage Groups used in the Dietary Reference Intakes (Institute of Medicine (IOM), 1997) excluding children under 4 years, fasters, and pregnant and lactating women. Values were controlled for total energy intake during the analyses.
^bTotal choline values available for NHANES 2005–2006 only and sample sizes for choline are shown in parentheses. Total choline content was calculated as the sum of free choline (Cho); lycerophosphocholine (GPC); phosphocholine (Pcho); phosphatidylcholine (PTdcho); and sphingomyelin (SM) (Patterson et al, 2008). Sample sizes for total choline were not adequate to generate percent greater than the AI.
^cOur analyses employed two days of 24-hour dietary recall data from the National Health and Nutrition Examination Survey (NHANES), What We Eat in America 2003–2006. We restricted our sample to children and adults as defined by the DRIs as aged 4 years or older. Our sample was further restricted to individuals with reliable recall status, excluding fasters, pregnant women, and lactating women. The final analytic sample included 29,099 days of recall data from 15,189 individual respondents. For 2,539 individuals in the sample, only one 24-hour recall was reported.
^dEAR or AI for the respective life stage group.
^eThe NHANES provides values for folic acid and food folate in Dietary Folate Equivalents (DFE). We present folate as Dietary Folate Equivalents. IDFE = 1 μg food folate = 0.6 mg of folic acid from fortified food.

Across the 13 nutrients plus fiber presented here, children 4 to 8 y were at risk of inadequacy for fewer nutrients than children 9 to 18 y when compared with the appropriate DRIs. Among children 4 to 8 y, in the lowest category of added sugars intake (⩽15% of energy from added sugar), vitamin E was the only nutrient for which a large proportion of children (42%) had an estimated median usual intake lower than the EAR (other nutrients: 0 to 6%). Older children, ages 9 to 18 y, who consumed ⩽15% of energy from added sugars had a high proportion of individuals with intakes below the EAR for both vitamin E (71 to 97%) and vitamin A (20 to 48%). As added sugars intake increased to >25%, 20% or more of children 4 to 8 years additionally had estimated usual intakes lower than the EAR for vitamin A. All adult life stage groups in the lowest added sugars category (<15% of energy) had more than 30 percent of individuals with intakes below the EAR across the age range 19 to 71+ y for vitamin A (range across age/gender groups: 38 to 47%), vitamin E (86 to 97%), vitamin C (31 to 39%), and magnesium (45 to 57%). For these nutrients, a very high percent of individuals across the life stage groups exhibited intakes below the EAR in the highest added sugars category (>25% of energy) (range across age/gender groups: vitamin A: 69 to 75%; vitamin E: 97 to 100%; vitamin C: 58 to 66%; magnesium: 83 to 90%).

Relatively few 9 to 13 y males and females and 14 to 18 y males had folate intakes below the EAR in the lowest added sugars category (9 to 13 years: males: 2%; females 11%; 14 to 18 year males: 12%); however, the percent of children with intakes below the EAR was considerably higher in the lowest added sugars category for females 14 to 18 years (36%) and in the highest added sugars category for males 14 to 18 y and females 9+ y (42, 40, and 72%, respectively). Among adults for folate in the lowest energy category (<15% of energy), fewer males than females in each life stage group had intakes below the EAR (range across male age groups: 11 to 12%; female age groups: 34 to 36%), with the percent of individuals who had intakes below the EAR more than doubling for both males and females in the highest added sugars category (range across male age groups: 39 to 41%; female age groups: 71 to 73%).

For sodium, 94% or more of individuals in all child and adult DRI life stage groups consistently met or exceeded the AI. We did not assess the percent of individuals in the three added sugars groups for intakes above the UL for all nutrients; however, children in all life stage groups from 4 to 18 years across all added sugars groups had median sodium intakes that exceeded the UL for their respective DRI groups (4 to 8 years: 1.9 g/day; males and females 9 to 13 years: 2.2 g/day; males and females 14 to 18 years: 2.3 g/day). Adults in all life stage groups across all added sugars groups, except males and females 71+ years who consumed >25% of energy from added sugars, had intakes of sodium that exceeded the UL of 2.3 g. Note that the sample sizes for adults 71+ years are very small in the highest added sugars category. The median nutrient values for this age/category for males and females should therefore be viewed accordingly. As with other nutrients, the lowest median sodium intake was among individuals who consumed >25% of energy from added sugars and thereby had sodium intakes closer to the UL.

Few children 4 to 8 y had intakes greater than or equal to the AI for potassium and fiber even in the lowest added sugars intake category (6% and 4%, respectively). Similar to the younger children, few older children (9+ y) exhibited intakes at or above the AI for potassium (0 to 4% across the life stage groups) and fiber (0 to 2% across the life stage groups). Similar to children, few adults exceeded the AI for fiber across all adult life stage groups (0 to 7%), even with added sugars intake <15% of energy.

For total choline we could not assess intake levels relative to the AI for each life stage group due to small sample sizes. However, for all life stage groups the nutrient intake was less with increased added sugars intake category from the lowest to the highest. Among children and adults, the median total choline intake was higher among males 9+ y than in females. These subgroup gender differences decreased when total energy intake was a covariate in the analysis (Appendix Tables 5 and 6), indicating that a portion of this gender difference in the total choline intake was attributable to differences in total energy intake across males and females.

In general, comparison of the life stage group intakes in Tables 4 and 5 with the life stage intakes in Appendix Tables 5 and 6 (where the total energy was a covariate in the analysis), shows that controlling for total energy in the analysis yields relatively small differences in the estimated percent of each life stage group above or below the relevant DRI critical values; however, the degree of difference between the energy-adjusted and non-energy-adjusted estimates varies by specific life stage and nutrient. In both the energy-adjusted and non-energy-adjusted estimates, all life stage intakes were less with increased added sugars intake, with the lowest nutrient intakes among persons with added sugars intake >25% of energy.

Food Sources

The mean intakes of added sugars in gram-equivalents from the select food categories for all age groups 4+ years combined not controlling for total energy in the analysis are shown in Table 6. On average, individuals in the U.S. ingested 83.0 gram equivalents (g-eq) per day of added sugars. The majority of these added sugars were consumed from grain products (18.7 g-eq/day), primarily sweetened grains (10.5 g-eq/day), and sugar, sweets, and beverages (50.8 g-eq/day), primarily comprised of regular soft drinks (25.5 g-eq/day), sugar/sweets (11.4 g-eq/day), and regular fruitades/drinks (8.6 g-eq/day), with regular soft drinks having the highest approximate contribution to added sugars intake (30.7%). Together, regular soft drinks, sweetened grains, sugar/sweets and regular fruitades/drinks contributed an estimated 56 g-eq of added sugars or approximately 67.3% of the added sugars to the diet. If milk and milk products are added (8.2 g-eq/d) the total contribution is approximately 77.2%. ¹⁰ Data from an earlier study of CSFII 1994–1996 are included in Table 6 (Guthrie and Morton, 2000).

Table 6 Mean Intake of Added Sugars in Gram-Equivalents [g-eq (SEM)] ^a by Food Category for all people aged 4 years and above by Range of Percent of Estimated Daily Intake from Added Sugars, National Health and Nutrition Examination Survey (NHANES) 2003–2006 ^b and from CSFII 1994–996 ^c

Food Category	CSFII 1994–1996^c		This Study^b
	15,010		10,432
		Approximate Contribution		Approximate Contribution
n	Mean g-eg^a	to Added Sugars Intake (%)	Mean g-eg (SEM)^a,e	to Added Sugars Intake (%)
Total added sugar consumed	82.2		83.0 (0.05)
Milk and milk products	7.1	8.6	8.2 (0.24)	9.9
Meat/poultry/fish/eggs/legumes	1.7^c	1.7	1.1 (0.04)	1.9
Grain products^f			18.7 (0.41)
Breakfast cereal	3.6	4.4	3.6 (0.13)	4.3
Sweetened grains	10.6	12.9	10.5 (0.32)	12.6
Other grains	4.8	5.8	4.4 (0.09)	5.3
Fruits and fruit products	1.1	1.3	0.8 (0.07)	1.0
Vegetables and vegetable products	1.1	1.3	0.9 (0.05)	1.1
Fats/oils/salad dressings	0.7	0.9	0.6 (0.04)	0.7
Sugars/sweets/beverages^g			50.8 (1.13)
Sugars/sweets	13.2	16.1	11.4 (0.33)	13.7
Regular soft drinks	27.1	33.0	25.5 (0.88)	30.7
Regular fruitades/ fruit drinks	8.0	9.7	8.6 (0.34)	10.3
Low calorie beverages	0.1	0.1	0.02 (0.00)	0.0
Other beverages	3.0	3.6	4.2 (0.29)	5.1
^aOne gram–equivalent equals an amount of added sugar comparable to 1 g sucrose in carbohydrate content.
^bSample excludes under 4 years, fasters, and pregnant and lactating women. Analysis controlled for total energy intake. Mean gram-equivalents of added sugar data from Guthrie and Morton (2000). Sample included all individuals 2 years of age and older, excluding breast-fed infants. Data were analyzed using USDA survey weights throughout. Meat and poultry food category included legumes in this study. No overall total for grains and sugars, sweets and sweetened beverages were included in this study. Alcoholic beverages were in the same category as other beverages in this study. Amounts in this table were estimates using NHANES 2003–2006 data and the NCI usual intake method. Total energy intake was not included as a covariate in the analysis that produced this data for better comparison with Guthrie & Morton (2000).
^cMean gram-equivalents of added sugar data from Guthrie and Morton (2000). Sample included all individuals 2 years of age and older, excluding breast-fed infants. Data were analyzed using USDA survey weights throughout. Meat and poultry food category included legumes in this study. No overall total for grains and sugars, sweets and sweetened beverages were included in this study. Alcoholic beverages were in the same category as other beverages in this study.
^dTotal may not add to 100 due to rounding and weighting of data.
^eThe standard errors reported in this table have been adjusted via balanced repeated replication, and therefore account for the complex NHANES sampling design.
^fExamples of grain products sub-categories include: breakfast cereals: sweetened and unsweetened cereals, cooked cereals, rice cereals, cereals not cooked or unknown as to cooked; sweetened grains: cakes, cookies, pies, pastries, crackers and salty snacks from grains, funnel cakes; other grains: flour and dry mixes, muffins, yeast breads, yeast rolls, quick breads, pancakes, waffles, French toast, crepes, four-water patties, flour-milk patties, rice flour cakes, grain mixtures, meat substitutes that are mail cereal protein.
^gExamples of sugars/sweets/beverages sub-categories include: sugars/sweets: sugar, sugar substitutes, candy, ice cream toppings, syrups, chewing gum, gum drops, ice pops, gelatin desserts, jams, jellies; regular soft drinks: cola and non-cola type, cream sodas. Ginger ales, root beer, chocolate-flavored soda; regular fruitades/fruit drinks: all single fruit and mixed fruit beverages such as apple juice, apple-cranberry-grape juice, fruit punch, lemonade, fruit flavored sports drinks; fruit flavored powdered drinks; low calorie beverages: any sugar-free or low calories beverage, carbonated or not carbonated; other beverages includes: alcoholic beverages: any alcoholic drinks including beer, wine, cocktails; other non-alcoholic beverages: coffee, tea, coffee substitutes, non-fruit beverages, non-alcoholic beer, wines, cocktails, non-carbonated water.

Table 7 shows the mean intakes of added sugars in g-eq from the select food categories for all age groups 4+ years combined for the eight added sugars categories. Total energy intake and data from the NHANES FFQ were included as covariates in this analysis. These data illustrate that for the major food sources of added sugars in the U.S. diet (milk and milk products, sweetened grains, sugars/sweets, regular soft drinks, regular fruitades/ fruit drinks, other non-alcoholic beverages) the estimated mean g-eq intake of added sugars from these sources increased monotonically across the eight added sugars categories (from 0 to >35% added sugars intake). In contrast, across the eight added sugars intake categories, the estimated added sugars contribution from other food sources remained relatively consistent. Appendix Table 7 presents these food category source data by DRI life stage groups.

Table 7 Mean Intake of Added Sugars in Gram-Equivalents [g-eq (SEM)] ^a by Food Category for all people aged 4 years and above by Range of Percent of Estimated Daily Intake from Added Sugars ^b With Total Energy Intake as a Covariate and Covariates from the Food Frequency Questionnaire in the Analysis, National Health and Nutrition Examination Survey (NHANES) 2003–2006

	Two-Day Mean of Energy from Added Sugars
Food Category	0 ⩽ x ⩽ 5%	5 < x ⩽ 10%	10 < x ⩽ 15%	15 < x ⩽ 20%	20 < x ⩽ 25%	25 < x ⩽ 30%	30 < x ⩽ 35%	>35%
N	987	1,976	2,551	2,278	1,427	689	275	249
Mean g-eg (SEM)
Total added sugar consumed	24.9 (0.54)	45.0 (0.79)	67.0 (0.95)	91.6 (1.03)	127.2 (1.74)	160.4 (3.31)	197.6 (4.95)	232.9 (7.48)
Milk and milk products	4.6 (0.24)	6.1 (0.24)	7.8 (0.25)	8.9 (0.30)	10.9 (0.44)	12.2 (0.61)	13.7 (0.84)	15.6 (1.15)
Meat/poultry/fish/eggs	0.9 (0.05)	1.1 (0.05)	1.1 (0.04)	1.1 (0.04)	1.3 (0.05)	1.3 (0.06)	1.4 (0.09)	1.4 (0.10)
Legumes	0.5 (0.04)	0.5 (0.06)	0.5 (0.05)	0.6 (0.06)	0.6 (0.08)	0.7 (0.10)	0.6 (0.12)	0.6 (0.11)
Grain products^c,d	11.9 (0.34)	15.6 (0.40)	18.5 (0.46)	20.2 (0.52)	23.2 (0.68)	24.8 (0.98)	26.4 (1.61)	25.0 (1.63)
Breakfast cereal	2.2 (0.11)	2.9 (0.13)	3.5 (0.14)	4.0 (0.19)	4.5 (0.19)	4.5 (0.29)	4.6 (0.38)	4.6 (0.34)
Sweetened grains	5.6 (0.29)	8.0 (0.36)	9.9 (0.35)	11.4 (0.36)	14.5 (0.59)	16.6 (0.90)	19.0 (1.53)	19.8 (1.76)
Other grains	3.8 (0.09)	4.2 (0.08)	4.4 (0.10)	4.4 (0.10)	4.8 (0.15)	4.8 (0.14)	4.8 (0.28)	4.5 (0.31)
Fruits and fruit products	0.6 (0.07)	0.7 (0.07)	0.8 (0.09)	0.8 (0.09)	0.9 (0.09)	0.9 (0.12)	1.1 (0.21)	1.0 (0.14)
Vegetables and vegetable products	0.8 (0.07)	0.9 (0.06)	0.7 (0.07)	0.9 (0.06)	1.0 (0.07)	1.0 (0.08)	0.9 (0.10)	0.9 (0.11)
Fats/oils/salad dressings	0.8 (0.07)	0.7 (0.05)	0.6 (0.04)	0.5 (0.03)	0.5 (0.04)	0.4 (0.04)	0.4 (0.06)	0.3 (0.04)
Sugars/sweets/beverages^c,e	7.7 (1.01)	19.8 (0.58)	36.5 (0.98)	57.0 (1.30)	86.0 (2.43)	115.5 (2.42)	152.5 (10.03)	187.6 (14.52)
Sugars/sweets	4.9 (0.20)	7.4 (0.24)	10.0 (0.27)	12.7 (0.37)	16.7 (0.66)	20.7 (0.82)	24.1 (1.25)	27.4 (1.93)
Regular soft drinks	2.7 (0.19)	6.8 (0.30)	14.6 (0.50)	27.3 (0.80)	47.1 (1.06)	69.0 (2.42)	96.0 (3.93)	116.7 (4.12)
Regular fruitades/ fruit drinks	1.9 (0.15)	3.6 (0.22)	5.8 (0.23)	9.0 (0.37)	14.4 (0.86)	20.7 (1.18)	27.2 (2.07)	37.4 (3.25)
Low calorie beverages	0.0 (0.01)	0.0 (0.00)	0.0 (0.00)	0.0 (0.00)	0.0 (0.00)	0.0 (0.01)	0.0 (0.00)	0.0 (0.00)
Alcoholic beverages	0.4 (0.10)	0.5 (0.10)	0.5 (0.06)	0.5 (0.05)	0.5 (0.08)	0.6 (0.14)	0.6 (0.14)	0.6 (0.20)
Other non-alcoholic beverages	1.2 (0.10)	1.8 (0.15)	2.6 (0.17)	3.6 (0.25)	5.7 (0.53)	7.8 (0.72)	9.6 (0.91)	15.1 (2.48)
^aOne Gram –equivalent equals an amount of added sugar comparable to 1 g sucrose in carbohydrate content.
^bSample excludes under 4 years, fasters, and pregnant and lactating women. Analysis controlled for total energy intake.
^cMean gram-equivalents of added sugar data from Guthrie and Morton (2000). Sample included all individuals 2 years of age and older, excluding breast-fed infants. Data were analyzed using USDA survey weights throughout. Meat and poultry food category included legumes in this study. No overall total for grains and sugars, sweets and sweetened beverages were included in this study. Alcoholic beverages were in the same category as other beverages in this study.
^dExamples of grain products sub-categories include: breakfast cereals: sweetened and unsweetened cereals, cooked cereals, rice cereals, cereals not cooked or unknown as to cooked; sweetened grains: cakes, cookies, pies, pastries, crackers and salty snacks from grains, funnel cakes; other grains: flour and dry mixes, muffins, yeast breads, yeast rolls, quick breads, pancakes, waffles, French toast, crepes, four-water patties, flour-milk patties, rice flour cakes, grain mixtures, meat substitutes that are mail cereal protein.
^eExamples of sugars/sweets/beverages sub-categories include: sugars/sweets: sugar, sugar substitutes, candy, ice cream toppings, syrups, chewing gum, gum drops, ice pops, gelatin desserts, jams, jellies; regular soft drinks: cola and non-cola type, cream sodas. Ginger ales, root beer, chocolate-flavored soda; regular fruitades/fruit drinks: all single fruit and mixed fruit beverages such as apple juice, apple-cranberry-grape juice, fruit punch, lemonade, fruit flavored sports drinks; fruit flavored powdered drinks; low calorie beverages: any sugar-free or low calories beverage, carbonated or not carbonated; alcoholic beverages: any alcoholic drinks including beer, wine, cocktails; other non-alcoholic beverages: coffee, tea, coffee substitutes, non-fruit beverages, non-alcoholic beer, wines, cocktails, non-carbonated water.

Methods Considerations

Total Energy Intake

To assess the impact of controlling for total energy intake when estimating usual nutrient intake, Appendix Table 4 includes the usual intake estimates for the DRI life stage groups and nutrients, including the total energy intake as a covariate in the usual intake calculation (comparable to Table 2 without energy as a covariate). Appendix Table 4 presents the eight added sugars intake categories for nutrients for all DRI life stage groups combined, without the total energy as a covariate (comparable to Table 3 with energy as a covariate). Controlling for the total energy intake reduced the overall variability in estimated nutrient intake levels across DRI life stage groups, although the magnitude of this effect varied by specific nutrient and life stage group. Comparing the values in Table 2 (total energy not a covariate) and Appendix Table 3 (total energy as a covariate), the difference in estimated median intake per life stage group per nutrient ranged from 0.0 to 16.6%.

The NCI versus ISU Method for Usual Intake Estimation

We generated estimates of the usual nutrient intake by DRI life stage groups using data from NHANES 2001–2002, for comparison with previously-published 2001–2002 estimates obtained using the ISU method (Moshfegh et al., 2005). ¹¹ Appendix Table 1 provides our estimates of the 2001–2002 nutrient intake data using the NCI method with total energy as a covariate; Appendix Table 2 provides the nutrient intake estimates calculated without total energy as a covariate. Appendix Table 2 results are comparable to the estimates from the prior USDA study, which did not control for total energy intake. In general, the usual nutrient intake estimates obtained using the NCI method for 2001–2002 were similar to those from the prior USDA study using the ISU method. ¹² The inclusion of the total energy as a covariate in Appendix Table 1 reduced the overall variability across life stage groups.

DISCUSSION

Dietary intake in the United States across all DRI life stage groups is not meeting the DRI recommendations for the majority of the 13 nutrients plus fiber in this study. For five of the eight nutrients with an EAR, few life stage groups had less than 10% of their median intakes below the EAR. For many adults, for these nutrients well over 50% of the life stage group had intakes below the EAR. In contrast, for all life stage groups, at least 94% of individuals met or exceeded the AI for sodium, and the median for all adult life stage groups exceeded their respective UL. The total fiber intake estimates were very low in children and adults with only older women (51+ years) evidencing over 10% of their life stage groups exceeding the AI. All other life stage groups ranged from 0 to 5% of fiber intake equal to/above the AI. The estimated total fiber intake in grams was lower in women than men for all life stage groups with or without inclusion of the total energy intake was a covariate in the analysis.

Similar to a recent report on usual nutrient intake from diet based on the NHANES 2001–2002 (Moshfegh et al., 2005), our data show that the U.S. population across 13 DRI life stage groups remains at the greatest risk for nutrient inadequacy from diet alone for vitamins E, A, and C and magnesium when median intakes are compared with the EAR for these nutrients. For other nutrients specific life stage groups are at higher risk of inadequacy; for example: females 9 to 18 y for phosphorus and zinc; females 9+ y for folate. In contrast to the 2001–2002 report, our estimates for iron indicated that less than 10% of females 14 to 50 years had median intake levels below the EAR, and therefore these data may indicate a slight improvement in iron status for these life stage groups.

While the AI cannot be used to estimate prevalence of inadequacy at the population level (IOM, 2000a), the proportion of the population above the AI can provide insight into general nutrient intake patterns, especially when only a small percent of the population achieves the AI. In our study, for potassium and fiber, 0 to 6% of individuals across all life stage groups had median intakes above the AI, with the exception of older females (51 to 71+ years: 13 to 15% above the AI for fiber). Similar to the 2001–2002 data, calcium and vitamin K are also of concern with an average of only 28% and 30% of the individuals, respectively, across the life stage groups having intake levels above the AIs. Approximately half as many females as males 9 to 71+ years achieve the AI intake recommendations for calcium.

It is in this context of low micronutrient intake in general, that we divided the NHANES study population into eight categories based on 5% increments in energy intake from added sugars to assess the impact of added sugars on nutrient intake. Over 87% of the U.S. population were estimated to have intake levels of added sugars between 0 and <25% of their total energy intake. The percent of the population was 6.6% in the > 25 to ⩽30% of added sugars intake category, and then was approximately 3% in each category for added sugars 30 to ⩽35 and >35%, respectively. Males and females did not differ in their distribution across the eight added sugars categories. However, relatively more non-Hispanic Black and individuals with incomes below the poverty line were associated with the highest level of added sugars intake than the overall population. Studies that demonstrate an increased risk of nutrient inadequacy for these racial and demographic groups are well known and recent research continues to underscore the need for low income and ethnic-specific interventions (Kant and Graubard, 2007).

A greater proportion of individuals classified as underweight and normal weight by BMI, reported higher levels of added sugars intake than individuals classified as overweight or obese. A recent review on carbohydrate intake and BMI reported that a number of epidemiological studies have shown this inverse relationship between BMI and carbohydrate intake (Gaesser, 2007). This review reported that studies it cited had BMI differences that were larger than what would be expected due to the reporting error alone (Gaesser, 2007). In addition, several of the studies in the review, similar to the NHANES data in our study, used the U.S. Department of Agriculture automated multiple pass (AMPM) 24-hour recall method (Raper et al., 2004), where the degree of underreporting has been assessed for normal, overweight, and obese individuals (Conway et al., 2003; 2004). However, underreporting of food intake also has been reported as inversely related to BMI (Subar et al., 2003). A recent evaluation of the accuracy of the AMPM method that compared self-reported energy intake to total energy expenditure using the doubly labeled water technique found that overall, the study participants underreported energy intake by 11%, but that underreporting was highest for overweight and obese participants (Moshfegh et al., 2008). Therefore, the differences in added sugars intake related to BMI that we observed may have been confounded by underreporting on the AMPM in the NHANES for overweight and obese individuals.

Similar to an earlier study (Bowman, 1999), for all life stage groups for all nutrients in general, the median estimated nutrient intake was lower with increased added sugars intake across the eight added sugars categories and was lowest for categories with energy from added sugars greater than 25% of total energy. The earlier IOM macronutrient report Appendix Table J analysis showed a similar trend for many nutrients; however, the data were more variable across the 5% increments in added sugars intake. For persons in the highest added sugars category, we found that the proportion of the life stage group with usual nutrient intake less than the EAR can double for some nutrients when compared to the estimated usual intake for the entire life stage group.

With 87% of the population having added sugars intake between 0 and 25% of energy from added sugars, what is the impact of added sugars on nutrient intake, particularly for those nutrients where the risk of inadequacy is the greatest? For all nutrients our data showed that median intakes exhibited a monotonically decreasing pattern (with lower median intake levels with each 5% increase in added sugars intake category), and the median nutrient intake in the highest added sugars category (>35% of energy) with all life stage groups combined, were 40% or more less than the median intakes for the 0 to 5% added sugars category. For vitamins E, A, C, and magnesium comparing our data with the total energy as a covariate, the percent of individuals whose intake was less than the EAR was higher with increased added sugars intake. However, for each nutrient, the relationship between the nutrient intake at 0 to15% of energy from added sugars to the general intake level for that life stage group and the magnitude of the difference between the lowest added sugars category and the highest differed. For vitamin E, where typically 80% or more of the population across the life stage groups had basic intakes below the EAR, the lowest added sugars category typically had 80+% of individuals below the EAR with the highest added sugars category including approximately 90 to 99% of persons with intakes less than the EAR. For vitamins A, C, and magnesium, fewer individuals with added sugars intakes in the lowest added sugars category (0 to 15% of energy) had nutrient intakes below the EAR than in the general population and the percent doubled or more for those in the highest added sugars intake category.

These data indicated that for all nutrients, but especially for the nutrients for which the U.S. population is most at risk for inadequate intake, the suggested maximal limit of 25% of energy from added sugars may need to be revisited. The added sugars intakes in even the 0 to 15% intake range may need further consideration for high risk nutrients for specific life stage groups. However, will readjusting the suggested maximal limit of added sugars downward be meaningful for those nutrients where the intake is already broadly inadequate based on the EAR? We believe any evaluation first should consider the questions: What magnitude change in percent life stage population below the EAR is meaningful for a 5 percent change in added sugars intake? What percent of a life stage group below the EAR constitutes a serious health concern for the nation?

As described in an earlier report based on the NHANES 2001–2002 (Moshfegh et al., 2005), the total fiber intake in the U.S. is particularly low. Our data extended this finding to the most recent NHANES and indicated that with higher added sugars intake the total fiber intake is lower still for all life stage groups. Epidemiological and intervention studies have demonstrated the positive health results that accompany dietary fiber intake particularly related to cardiovascular disease (CVD) and to some degrees for diabetes (Mozaffarian et al., 2003; Venn and Mann, 2004; Erkkilä and Lichtenstein, 2006; Mann, 2007; Priebe et al., 2008; Theuwissen and Mensink, 2008). The IOM macronutrient report based their recommendations for the AI for total fiber in part on literature linking fiber and CVD and diabetes (IOM, 2002b). The Dietary Guidelines (DHHS/USDA, 2005) emphasized the selection of fiber-rich foods, and provided clear information on the multiple health values of a fiber-rich diet. However, clearly more public health education is needed to address the low fiber intake of our population.

We found that the overall food category sources of added sugars were similar to earlier studies (Bowman, 1999; Guthrie and Morton, 2000). In the 1994–1996 CSFII analysis, the four food categories of regular soft drinks, sugar/sweets, sweetened grains, and regular fruitades/drinks comprised 72% of the intake of added sugars. In our study, the same four food categories comprised approximately 67% of the added sugars in the diet.

Earlier analysis of the 1994–1996 CSFII found that the mean intake of total added sugars for the U.S. population 2 years and older (15,010 individuals) was 82.2 gram-equivalents (Guthrie and Morton, 2000). This amount is remarkably similar to the 83.0 mean gram-equivalents we calculated for the U.S. population 4 y and older (10,432 individuals) based on the NHANES 2003–2006 data. In Table 6 we included the mean gram-equivalent intakes of added sugars from 13 food categories from the CSFII study for comparison with our data. It is important to consider the differences in methodology between the earlier study and our own in reviewing these data. The CSFII study used one day of intake data for persons 2 years and older, did not estimate the usual intake, and used USDA survey weights appropriate for the study and time frame. Both studies used the USDA Food Guide Pyramid Servings Database definition of added sugars and teaspoon values for added sugars in the database to convert to gram-equivalents added sugars. In our study we used similar food categories and subcategories as the CSFII study. Both the CSFII and our weighted NHANES 2003–2006 data thus can be interpreted in terms of the entire U.S. population at the time period the surveys were conducted, but our methodology included usual intake estimates based on the USDA validated AMPM data collection method currently used in NHANES. The resulting data from our study took into consideration additional variability in intake in terms of day-to-day and food type variation at the individual level.

Overall, the resulting estimated mean gram-equivalent intake by food category for the populations differed very little between the two studies. In fact, the mean gram-equivalent intake estimates for grains differed by tenths of a gram. The predominant sources of added sugars intake were consistent for the two studies with the highest gram-equivalent intake of added sugars from regular soft drinks (CSFII 1994–1996: 27.1 g-eq/day; NHANES 2003–2006: 25.5 g-eq/day), sugars and sweets (CSFII 1994–1996: 13.2 g-eq/day; NHANES 2003–2006: 11.4 g-eq/day), and sweetened grains (CSFII 1994–1996:10.6 g-eq/day; NHANES 2003–2006: 10.5 g-eq/day).

Due to methodological difference, we cannot compare life stage mean intakes between the two studies. In addition, the CSFII study used age/gender groups based on the 1989 Recommended Dietary Allowances (NRC, 1989a) which did not match the age ranges in the DRI life stage groups, which we used for this study, for example, 1989 RDA: 18 to 34 years versus DRI: 19 to 30 years. However, similar trends between the two studies were evident across the age/gender groups. For example, regular soft drinks were the main contributors of added sugars to the diet for all age/gender groups with a higher mean gram-equivalent contribution of added sugars from soft drinks beginning in adolescence, peaking in adulthood, and evidencing lower contribution after age 50 (See Appendix Table 7). Among grain products in both studies, while sweetened grains were the highest contributors of gram-equivalents of added sugars, added sugars intake from breakfast cereals was age-dependent with the highest added sugars intake from this food category among children less than 18 years and lower intakes in adults, particularly the oldest age groups. In general, the data from these two nationally representative surveys, even considering the changes in methodology, indicated that the amount in mean gram-equivalents of added sugars per day and food category sources of added sugars has not changed appreciably between the two study periods.

In Bowman's study that similarly evaluated added sugars intake using the 1994–1996 CSFII data, the population was divided into three added sugars categories based on—<10%, ⩾10 ⩽18%, >18% of energy from added sugar (Bowman, 1999). This author included estimated mean (± SE) energy intakes for these three groups of—<10%: 1860 ± 15.0 kcal; ⩾10 ⩽18%: 2040 ± 18.1 kcal; >18%: 2049 ± 17.2 kcal. While the eight added sugar groupings in our study did not match those of the earlier study, in our study the median energy intake overall was 2118 ± 13.1 kcal with the median energy intake for the 2003–2006 NHANES sample being considerably higher in the added sugar categories than the earlier CSFII numbers. (See Table 1.) For example, in the current study the median total energy intake estimates at the two lowest end of the added sugar categories for ⩾0 ⩽ 5% and ⩾5 ⩽10% energy from added sugars were 2028 ± 17.4 kcal and 2063 ± 15.4 kcal, respectively, compared with a mean of 1860 kcal for the <10% added sugars category in the 1994–1996 CSFII data. In the highest added sugar category of the CSFII data, this author reported a mean total energy intake of 2049 ± 17.2 kcal (>18% energy from added sugars). In the present study, the estimated mean total median daily energy intake for the >20 ⩽25% added sugars category was 2177 ± 15.2 kcal with the >35% added sugars category total median daily energy intake estimate being 2298 ± 28.3 kcal. Therefore, keeping data collection and the analysis differences in mind, it appears that the overall energy intake clearly has increased while the estimated added sugar intake in g-eg/day has remained relatively constant between the 1994–1996 CSFII to the 2003–2006 NHANES nationally representative estimates.

We found that controlling for total energy intake changed the estimated nutrient intake from 0.0 to 16.6% in a variable fashion that was specific to the nutrient and the life stage group. Our results therefore support earlier contentions (Forshee and Storey, 2004; 2005; Barr and Johnson, 2005) that patterns of estimated nutrient intake by added sugars category in the Table J Appendix of the IOM macronutrient report would have differed if total energy intake had been considered. Specifically, it appears that failing to adjust for total energy intake may in part mask differences between added sugars categories for most nutrients, in that the nutrient intake differs to a greater degree across the added sugars categories when the total energy intake is included as a covariate in the analysis.

In this study, we employed the new NCI method for estimating the usual intakes and used the NCI-supplied SAS macros that were first released in late 2008. The NCI method accounts for a possible correlation between the amount and the frequency of consumption, allows the incorporation of covariates, and permits efficient estimation of usual intake for subgroups. Our analyses exploited each of these novel modeling capabilities. In estimating the usual intake of added sugars by food source, we employed a two-part model, jointly estimating both the probability and the amount of consumption. Both our added sugars and our nutrient intake models included the total energy intake and the day of the week as model covariates, and our added sugars models additionally incorporated covariates from the NHANES 2003–2006 Food Frequency Questionnaire. Finally, since we estimated the usual intake for subgroups defined by DRI life stage and by added sugars intake, the efficiency gains from estimating the model for all subgroups simultaneously rather than by stratifying our sample were likely substantial.

Because the NCI method is so new we felt it essential to assess the results we obtained using this method against previously published data that used the ISU method. Therefore, we generated the usual intake estimates using the NCI method for the NHANES 2001–2002 nutrient intakes and compared the results with NHANES 2001–2002 results produced by USDA scientists using the ISU usual intake approach (Moshfegh et al., 2005). We found remarkable agreement among the NHANES 2001–2002 nutrient intake results using these two methods and thus concluded that we appropriately applied this new method.

FFQ data were not collected from NHANES respondents earlier than 2003–2004. Because we wished to incorporate NHANES FFQ data as a covariate into our added sugars by food source analyses, our sample size was limited to NHANES data since 2003. To ensure adequate sample size for our analyses, we incorporated NHANES 2005–2006 data, for which the corresponding MyPyramid Equivalent Database (MPED) added sugars data had not been released. As described above, we thus applied the MPED 2.0 data, which were developed for NHANES 2003–2004, to estimate individual added sugars intake from the NHANES 2005–2006 recalls, with a trained nutritionist supplying added sugars estimates for the 68 new foods consumed by NHANES respondents in 2005–2006. To the extent that standard MPED recipes change over time, our estimates may differ slightly from those that would have been obtained had the most up-to-date MPED data been available. The total choline in foods was first included for the NHANES 2005–2006 data; therefore our sample for this nutrient was approximately half that of the other nutrients. Due to low sample sizes for the total choline, another study limitation was that we could not estimate the proportion of individuals in each life stage/added sugars group with intake greater than or equal to the AI for this nutrient.

We have found, similar to the earlier report of nutrient intake from food based on the NHANES 2001–2002 (Moshfegh et al., 2005), that a large proportion of the U.S. population in general is at risk for nutrient inadequacy for many nutrients, based on the EAR. This high proportion of individuals with intakes below the EAR potentially places the population at risk for chronic health conditions that have been associated with these nutrients such as poor bone growth, osteoporosis, cardiac complications, neuromuscular excitability, poor regulation of blood pressure, etc. Our data also indicated that a higher overall proportion of non-Hispanic Blacks and persons with incomes below the poverty line were included in the highest added sugars intake category (>25% of energy from added sugars) and thereby at potentially greater risk for nutrient inadequacy. These same racial and demographic characteristics are associated with increased risk of cardiovascular disease and diabetes (Brancati et al., 2000; Cox et al., 2004; Vijan et al., 2005; Lancaster et al., 2006; Jen et al., 2007).

A recent review postulated that micronutrient inadequacy in general may have broader implications beyond the specific effects of the individual nutrients in that generalized micronutrient inadequacies may result in “metabolic disruption” and mitochondrial decay (Ames, 2006). Such decay is associated with aging and degenerative disease, including cancer and dementia, and may be the consequence of a triage allocation response to micronutrient inadequacy. The author cited the poor dietary intake of micronutrients in the NHANES 2001–2002 study (Moshfegh et al., 2005) and recommended dietary supplement intake to reduce the risk of micronutrient inadequacy and concomitant risk of disease.

Persons in the United States currently are urged to obtain their nutrients from a healthful diet and provided with carefully developed guidance on diet and physical activity (DHHS/USDA, 2005; 2005 Dietary Guidelines Advisory Committee, 2004). However, the results presented in our study indicate that people are still far from meeting this guidance from food alone. The combined estimates of nutrients from food, dietary supplements, including intake of added sugars, may indicate a better nutrient profile for that third to half of the U.S. population who uses supplements. This research and its careful interpretation in terms of nutrient bioavailability by various life stage groups is sorely needed.

Addressing inadequate nutrient intake in the U.S. population has been and will continue to be a priority. It is clear from this assessment of NHANES intake data and other studies over several decades, that efforts to improve the intake of specific micronutrients and fiber have not been entirely successful.

CONCLUSION

A large proportion of the U.S. population continues to be at potential risk for nutrient inadequacy based on the NHANES 2003–2006. Similar to the IOM macronutrient report Appendix Table J (IOM, 2002b) and other reports (Bowman, 1999), we found that individuals with intakes of energy >25% from added sugars appear to be at greater risk for nutrient inadequacy based on comparison with the DRIs. For nutrients where people in specific life stage groups are at highest risk, added sugars intake below 25% of energy may need to be considered. As in other studies, we identified regular soft drinks, sweets and sugars, and sweetened grains as the major contributors to added sugars intake on a gram-equivalent intake basis (Bowman, 1999; Guthrie and Morton, 2000; Vartanian et al., 2007). A comparison of gram-equivalent intakes of added sugars for the major food categories in this study with previous nationally representative data from 1994–1996 indicated little change in the estimated daily intake of added sugars both across the food categories and for trends in intake among life stage groups, while the estimated energy intake/day has increased. However, high levels of added sugars intake occur among only a small proportion of the population and cannot explain the existing problem of poor nutrient intake in the U.S. population as a whole. Since their inception, the Dietary Guidelines have provided thoughtful and carefully presented information for the U.S. population on food intake balance. This guidance needs to be more widely disseminated to reach the proportion of the population whose nutrient intakes fall outside expected ranges for nutrient adequacy, including non-Hispanic Blacks and persons with low income (King, 2007), and those life stage groups and sub-populations at most risk for specific nutrient inadequacy.

Appendix Table 1 Intake of Nutrients [Median (SE)] by Dietary Reference Intake (DRI) Life Stage Groups ^a With Total Energy (KCAL)as a Covariate in the Analysis, National Health and Nutrition Examination Survey (NHANES) 2001–2002

	DRI Life Stage Groups
	Children4 to 8 yrs	Males9 to 13 yrs	Females9 to 13 yrs	Males14 to 18 yrs	Females14 to 18 yrs	Males19 to 30 yrs	Females19 to 30 yrs	Males31 to 50 yrs	Females31 to 50 yrs	Males51 to 70 yrs	Females51 to 70 yrs	Males71+ yrs	Females71+ yrs
Nutrient
n	920	574	596	727	680	552	468	785	755	651	643	392	405
Median (SE)
Vitamin A, μg RAE	438 (1.0)	492 (1.6)	480 (1.5)	576 (2.0)	490 (1.5)	630 (2.5)	531 (1.9)	617 (1.9)	513 (1.6)	541 (3.0)	493 (1.4)	490 (1.7)	470 (1.6)
Vitamin E, mg	4.4 (0.01)	5.3 (0.02)	4.9 (0.01)	6.5 (0.03)	5.2 (0.02)	7.4 (0.02)	5.8 (0.02)	7.4 (0.02)	5.6 (0.02)	6.4 (0.06)	5.5 (0.01)	5.8 (0.02)	5.2 (0.02)
Vitamin C, mg	63 (0.2)	71 (0.3)	69 (0.3)	82 (0.3)	70 (0.3)	89 (0.4)	76 (0.3)	88 (0.3)	75 (0.3)	79 (0.4)	73(0.3)	72 (0.3)	70 (0.3)
Folate, μg^b	325 (9.6)	381 (17.7)	332 (10.8)	436 (16.8)	335 (14.4)	473 (20.3)	358 (6.9)	458 (14.5)	341 (10.2)	396 (10.5)	324 (8.8)	357 (10.0)	303 (7.6)
Iron, mg	12.2 (0.02)	14.5 (0.04)	12.2 (0.03)	16.7 (0.05)	12.3 (0.03)	18.0 (0.07)	13.1 (0.04)	17.4 (0.05)	12.4 (0.03)	14.9 (0.10)	11.7 (0.02)	13.3 (0.03)	10.9 (0.02)
Zinc,mg	9.0 (0.01)	11.0 (0.03)	8.9 (0.02)	12.8 (0.04)	9.0 (0.02)	14.0 (0.06)	9.7 (0.03)	13.5 (0.04)	9.1 (0.03)	11.5 (0.12)	8.6 (0.12)	10.2 (0.02)	7.9 (0.02)
Magnesium, mg	184 (0.3)	227 (0.6)	196 (0.4)	272 (0.8)	206 (0.5)	306 (1.0)	231 (0.61)	308 (0.70	228 (0.5)	275 (1.2)	223 (0.4)	256 (0.6)	217 (0.5)
Phosphorus, mg	1,067 (1.5)	1,275 (2.8)	1,070 (2.2)	1,464 (3.7)	1,076 (2.4)	1,591 (4.6)	1,159 (3.0)	1,531 (3.2)	1,080 (2.3)	1,295 (4.3)	1,004 (1.8)	1,143 (2.5)	914 (2.0)
Calcium, mg	798 (1.4)	884 (2.4)	797 (2.0)	982 (3.1)	769 (2.0)	1,025 (3.8)	788 (2.5)	946 (2.7)	711 (2.1)	765 (4.5)	636 (1.5)	644 (1.9)	559 (1.6)
Sodium, mg	2,974 (3.9)	3,534 (7.6)	2,989 (5.7)	4,046 (10.0)	3,014 (6.3)	4,401 (12.6)	3,245 (7.9)	4,250 (8.9)	3,043 (6.2)	3,627 (12.5)	2,851 (4.8)	3,233 (6.4)	2,622 (5.2)
Potassium, mg	1,836 (2.7)	2,271 (5.0)	1,944 (4.0)	2,703 (6.4)	2,052 (4.3)	3,022 (8.2)	2,294 (5.5)	3,048 (5.9)	2,275 (4.3)	2,757 (7.9)	2,244 (3.7)	2,594 (5.1)	2,189 (4.3)
Vitamin K, μg	33 (0.1)	40 (0.2)	43 (0.2)	54 (0.3)	48 (0.2)	66 (0.5)	58 (0.3)	71 (0.4)	62 (0.5)	66 (3.8)	65 (0.2)	65 (0.3)	67 (0.3)
Total Fiber, g	10 (0.02)	13 (0.03)	11 (0.03)	15 (0.04)	12 (0.03)	17 (0.06)	13 (0.04)	17 (0.04)	13 (0.03)	15 (0.06)	13 (0.03)	14 (0.04)	13 (0.03)
^aBased on the Dietary Reference Intake (DRI) Life Stage Groups used in the Dietary Reference Intakes (Institute of Medicine (IOM), 1997) excluding children under 4 years, fasters, and pregnant and lactating women. Values were controlled for total energy intake during the analyses.
^bThe NHANES provides values for folic acid and food folate in Dietary Folate Equivalents (DFE). We present folate as Dietary Folate Equivalents. IDFE = 1 μg food folate = 0.6 mg of folic acid from fortified food.

Appendix Table 2 Intake of Nutrients [Median (SE)] by Dietary Reference Intake (DRI) Life Stage Groups ^a Without Total Energy Intake as a covariate in the Analysis, National Health and Nutrition Examination Survey (NHANES) 2001–2002

	DRI Life Stage Groups
	Children4 to 8 yrs	Males9 to 13 yrs	Females9 to 13 yrs	Males14 to 18 yrs	Females14 to 18 yrs	Males19 to 30 yrs	Females19 to 30 yrs	Males31 to 50 yrs	Females31 to 50 yrs	Males51 to 70 yrs	Females51 to 70 yrs	Males71+ yrs	Females71+ yrs
Nutrient
n	920	574	596	727	680	552	468	785	755	651	643	392	405
Median (SE)
Vitamin A, μg RAE	520 (1.2)	573 (1.5)	494 (1.4)	579 (1.4)	498 (1.3)	583 (1.6)	507 (1.6)	594 (1.3)	519 (1.3)	602 (1.5)	528 (1.4)	615 (2.0)	538 (1.8)
Vitamin E, mg α-tocopherol	5.6 (0.01)	6.5 (0.01)	5.2 (0.01)	6.7 (0.01)	5.4 (0.01)	6.9 (0.01)	5.6 (0.01)	7.2 (0.01)	5.8 (0.01)	7.4 (0.01)	6.0 (0.01)	7.7 (0.02)	6.3 (0.02)
Vitamin C, mg	73 (0.2)	80 (0.3)	70 (0.3)	81 (0.3)	71 (0.2)	82 (0.3)	73 (0.3)	84 (0.3)	75 (0.2)	86 (0.3)	77 (0.3)	88 (0.4)	78 (0.3)
Folate, μg^b	389 (10.8)	445 (12.2)	344 (9.0)	445 (10.7)	343 (8.7)	443 (10.5)	345 (8.5)	446 (11.0)	348 (9.1)	446 (11.2)	349 (9.2)	449 (11.3)	350 (9.1)
Iron, mg	14.8 (0.02)	17.2 (0.03)	12.8 (0.02)	17.1 (0.03)	12.7(0.02)	17.0 (0.03)	12.7 (0.03)	17.0(0.02)	12.7 (0.02)	17.0 (0.03)	12.7 (0.02)	17.0 (0.03)	12.7 (0.03)
Zinc, mg	11.2 (0.02)	13.3 (0.02)	9.4 (0.02)	13.3 (0.02)	9.4 (0.02)	13.2 (0.02)	9.4 (0.02)	13.2 (0.02)	9.4 (0.01)	13.2 (0.02)	9.4 (0.02)	13.3 (0.03)	9.4 (0.02)
Magnesium, mg	224 (0.4)	267 (0.5)	204 (0.4)	277 (0.4)	211 (0.4)	286 (0.5)	220 (0.4)	298 (0.4)	230 (0.4)	309 (0.5)	240 (0.4)	321 (0.7)	250 (0.5)
Vitamin K, μg	412 (0.1)	49 (0.2)	45 (0.1)	55 (0.2)	50 (0.2)	61 (0.2)	56 (0.2)	69 (0.2)	63 (0.2)	77 (0.2)	71 (0.2)	87 (0.3)	80 (0.3)
Phosphorus, mg	1,298 (1.9)	1,503 (2.3)	1,111 (1.9)	1,492 (2.0)	1,099 (1.7)	1,476 (2.3)	1,095 (2.1)	1,472 (1.9)	1,094 (1.6)	1,462 (2.1)	1,088 (1.8)	1,459 (2.7)	1,083 (2.2)
Calcium, mg	966 (1.7)	1,043 (2.1)	825 (1.8)	997 (1.8)	784 (1.6)	950 (2.1)	751 (1.9)	913 (1.7)	723 (1.4)	873 (1.8)	690 (1.5)	839 (2.2)	661 (1.8)
Sodium, mg	3,034 (4.4)	3,530 (5.4)	2,598 (4.3)	3,510 (4.8)	2,575 (4.0)	3,481 (5.5)	2,571 (4.9)	3,478 (4.6)	2,576 (3.8)	3,463 (5.1)	2,567 (4.2)	3,463 (6.5)	2,563 (5.2)
Potassium, mg	2,196 (3.2)	2,627 (4.2)	2,005 (3.5)	2,727 (3.8)	2,083 (3.3)	2,822 (4.5)	2,180 (4.1)	2,940 (3.8)	2,287 (3.3)	3,051 (4.3)	2,385 (3.7)	3,178 (5.7)	2,490 (4.8)
Vitamin K, μg	412 (0.1)	49 (0.2)	45 (0.1)	55 (0.2)	50 (0.2)	61 (0.2)	56 (0.2)	69 (0.2)	63 (0.2)	77 (0.2)	71 (0.2)	87 (0.3)	80 (0.3)
Total Fiber,g	13 (0.02)	15 (0.03)	12 (0.02)	15 (0.03)	12 (0.02)	16 (0.03)	13 (0.03)	16 (0.03)	14 (0.02)	17 (0.03)	14 (0.03)	18 (0.04)	15 (0.03)
^aBased on the Dietary Reference Intake (DRI) Life Stage Groups used in the Dietary Reference Intakes (Institute of Medicine (IOM), 1997) excluding children under 4 years, fasters, and pregnant and lactating women. Values were NOT controlled for total energy intake during the analyses.
^bThe NHANES provides values for folic acid and food folate in Dietary Folate Equivalents (DFE). We present folate as Dietary Folate Equivalents. IDFE = 1 μg food folate = 0.6 mg of folic acid from fortified food.

Appendix Table 3 Estimated Usual Intakes of Nutrients from Food by Dietary Reference Intake (DRI) Life Stage Groups ^a Including Total Energy as a Covariate in the Analysis for Vitamin A, Vitamin E, Vitamin C, Folate, Iron, Zinc, Magnesium, Phosphorus, Calcium, Sodium, Potassium, Vitamin K, Fiber and Total Choline ^b , National Health and Nutrition Examination Survey (NHANES) 2003–2006, all groups combined

	DRI Life Stage Groups
	Children4 to 8 yrs	Males9 to 13 yrs	Females9 to 13 yrs	Males14 to 18 yrs	Females14 to 18 yrs	Males19 to 30 yrs	Females19 to 30 yrs	Males31 to 50 yrs	Females31 to 50 yrs	Males51 to 70 yrs	Females51 to 70 yrs	Males71+ yrs	Females71+ yrs
Nutrient
n (Total Choline n)^b	1,680(808)	1,009 (484)	1,039 (497)	1,351 (578)	1,250 (601)	1,097 (455)	914 (419)	1,439 (672)	1,350 (643)	1,215 (559)	1,251 (579)	808 (334)	787 (296)
Median (SE)
Vitamin A, RAE	442 (13.4)	489 (12.1)	479 (27.3)	584 (23.8)	498 (17.6)	627 (13.3)	513 (14.0)	636 (22.5)	509 (27.7)	551 (14.6)	490 (15.4)	494 (12.5)	470 (25.6)
Percent<EAR (μg/d)	19%	43%	40%	55%	48%	50%	48%	49%	49%	60%	52%	68%	55%
Vitamin E, mg α-tocopherol	4.6 (0.10)	5.4 (0.12)	5.1 (0.17)	6.8 (0.19)	5.4 (0.10)	7.5 (0.15)	5.7 (0.12)	7.8 (0.21)	5.7 (0.23)	6.6 (0.15)	5.5 (0.11)	5.9 (0.10)	5.3 (0.20)
Percent<EAR (mg/d)	75%	87%	90%	84%	96%	81%	94%	80%	95%	90%	97%	96%	98%
Vitamin C, mg	61 (1.8)	67 (1.8)	66 (1.5)	78 (2.1)	69 (1.4)	84 (2.1)	71 (1.9)	85 (2.0)	70 (1.9)	75 (2.1)	68 (2.0)	68 (2.1)	66 (2.3)
Percent<EAR [mg/d]	7%	21%	21%	37%	38%	43%	40%	43%	40%	50%	42%	56%	44%
Folate, μg^c	320 (4.4)	368 (6.4)	324 (12.5)	431 (12.7)	332 (5.6)	457 (7.0)	335 (5.2)	459 (14.5)	328 (14.6)	390 (9.0)	311 (6.2)	346 (6.1)	293 (12.5)
Percent<EAR [μg/d]	3%	15%	25%	27%	49%	20%	46%	19%	47%	31%	53%	42%	60%
Iron, mg	12.7 (0.17)	14.8 (0.24)	12.6 (0.42)	17.4 (0.46)	12.8 (0.19)	18.4 (0.28)	12.9 (0.20)	18.5 (0.53)	12.6 (0.53)	15.5 (0.35)	11.9 (0.21)	13.7 (0.25)	11.2 (0.48)
Percent<EAR (mg/d)	0%	1%	1%	2%	8%	0%	10%	0%	11%	0%	0%	1%	1%
Zinc, mg	9.4 (0.13)	11.2 (0.16)	9.2 (0.35)	13.4 (0.35)	9.4 (0.14)	14.3 (0.29)	9.5 (0.17)	14.3 (0.48)	9.3 (0.42)	11.9 (0.27)	8.8 (0.14)	10.4 (0.18)	8.2 (0.32)
Percent<EAR (mg/d)	0%	9%	21%	14%	23%	16%	18%	15%	18%	25%	21%	37%	26%
Magnesium, mg	191 (1.9)	230 (3.7)	203 (6.8)	283 (7.3)	217 (3.3)	311 (4.6)	228 (4.2)	325 (8.1)	233 (8.6)	286 (5.7)	229 (4.0)	263 (3.7)	225 (7.3)
Percent<EAR (mg/d)	7%	35%	48%	66%	81%	55%	61%	57%	64%	71%	67%	81%	71%
Phosphorus, mg	1,079 (14.3)	1,265 (18.2)	1,079 (42.9)	1,497 (36.2)	1,102 (16.9)	1,579 (27.0)	1,108 (20.5)	1,586 (40.3)	1,075 (43.3)	1,315 (22.6)	1,002 (15.8)	1,141 (18.8)	924 (29.4)
Percent<EAR [mg/d]	0%	30%	48%	21%	45%	1%	6%	1%	6%	2%	6%	3%	8%
Calcium, mg	844 (13.0)	914 (16.9)	843 (35.0)	1,041 (36.6)	829 (15.2)	1,066 (20.4)	804 (13.5)	1,031 (33.1)	752 (31.2)	824 (18.6)	678 (11.7)	687 (12.5)	603 (23.3)
Percent⩾AI (mg/d)	56%	20%	15%	33%	14%	56%	31%	53%	25%	20%	8%	9%	4%
Sodium, mg	2,770 (33.8)	3,244 (50.4)	2,773 (83.7)	3,841 (88.1)	2,840 (39.3)	4,055 (73.4)	2,860 (67.5)	4,084 (112.7)	2,784 (94.1)	3,397 (59.6)	2,608 (40.4)	2,968 (47.8)	2,420 (59.2)
Percent⩾AI [mg/d]	99%	99%	95%	99%	96%	99%	96%	99%	96%	100%	98%	100%	99%
Potassium, mg	1,842 (23.4)	2,229 (32.5)	1,945 (71.1)	2,715 (71.7)	2,078 (30.9)	2,970 (51.0)	2,191 (33.2)	3,097 (79.2)	2,236 (88.7)	2,753 (52.8)	2,212 (34.8)	2,555 (34.4)	2,174 (74.2)
Percent⩾AI [mg/d]	1%	3%	1%	10%	1%	13%	2%	12%	2%	6%	1%	2%	0%
Vitamin K, μg	40 (0.8)	47(1.1)	50 (1.7)	63 (2.0)	57 (1.5)	76 (2.8)	65 (2.2)	84 (2.5)	70 (1.9)	77 (3.1)	73 (2.9)	74 (3.3)	75 (1.7)
Percent⩾AI [μg/d]	27%	33%	37%	40%	31%	24%	28%	29%	32%	22%	34%	18%	36%
Total Fiber, g	10 (0.2)	12 (0.3)	11 (0.4)	15 (0.4)	12 (0.2)	17 (0.3)	13 (0.2)	17 (0.5)	13 (0.6)	15 (0.4)	13 (0.3)	14 (0.3)	13 (0.6)
Percent⩾AI [g/d]	1%	2%	2%	4%	3%	5%	5%	4%	5%	5%	9%	2%	7%
Total Choline, mg^b	211 (8.2)	265 (9.5)	220 (6.7)	333 (13.0)	233 (6.6)	363 (8.3)	238 (7.4)	389 (14.2)	255 (4.7)	357 (10.7)	250 (4.1)	324 (7.9)	249 (4.8)
^aBased on the Dietary Reference Intake Life Stage Groups used in the Dietary Reference Intakes (Institute of Medicine (IOM), 1997) excluding children under 4 years, fasters, and pregnant and lactating women. Values were NOT controlled for total energy intake during this analyses.
^bTotal choline values available for NHANES 2005–2006 only and sample sizes for choline are shown in parentheses. Total choline content was calculated as the sum of free choline (Cho); glycerophosphocholine (GPC); phosphocholine (Pcho); phosphatidylcholine (PTdcho); and sphingomyelin (SM) (Patterson et al., 2008). Sample sizes for total choline were not adequate to generate percent greater than the AI.
^cThe NHANES provides values for folic acid and food folate in Dietary Folate Equivalents (DFE). We present folate as Dietary Folate Equivalents. IDFE = 1 μg food folate = 0.6 mg of folic acid from fortified food.

Appendix Table 4 Estimated Usual Intakes of Nutrients from Food [Median (SE)] by Range of Percent of Estimated Daily Intake from Added Sugars Without Total Energy Intake as a Covariate in the Analysis for All Life Stages Combined ^a for Vitamin A, Vitamin E, Vitamin C, Folate, Iron, Zinc, Magnesium, Phosphorus, Calcium, Sodium, Potassium, Vitamin K, Fiber and Total Choline ^b , National Health and Nutrition Examination Survey (NHANES) 2003–2006.

	Categories of Individuals Based on Percent of Energy Intake from Added Sugars
Food Category	0 ⩽ x ⩽ 5%	5 < x ⩽ 10%	10 < x ⩽ 15%	15 < x ⩽ 20%	20 < x ⩽ 25%	25 < x ⩽ 30%	30 < x ⩽ 35%	>35%
N (Total Choline n)^b	1,387 (645)	2,877 (1,372)	3,666 (1,725)	3,311 (1,523)	2,055 (909)	1,078 (458)	432 (164)	384 (129)
Median (SE)
Vitamin A, RAE	652 (15.8)	609 (14.9)	570 (13.4)	530 (13.1)	494 (13.3)	461 (15.0)	429 (13.8)	397 (14.6)
Vitamin E, mg α-tocopherol	7.6 (0.10)	7.1 (0.08)	6.7 (0.06)	6.3 (0.07)	5.9 (0.07)	5.5 (0.08)	5.2 (0.09)	4.8 (0.10)
Vitamin C, mg	90 (2.5)	83 (1.8)	77 (1.5)	71 (1.4)	66 (1.4)	61 (1.7)	56 (1.9)	51 (2.2)
Folate, μg^c	442 (5.4)	417 (5.2)	394 (4.7)	370 (4.7)	349 (5.0)	330 (6.1)	310 (5.6)	291 (6.2)
Iron, mg	16.4 (0.18)	15.8 (0.17)	15.4 (0.15)	14.9 (0.17)	14.4 (0.19)	14.0 (0.25)	13.5 (0.25)	13.1 (0.28)
Zinc, mg	12.8 (0.14)	12.2 (0.14)	11.8 (0.13)	11.2 (0.15)	10.8 (0.18)	10.4 (0.22)	9.9 (0.23)	9.5 (0.26)
Magnesium, mg	317 (2.6)	295 (2.4)	275 (2.2)	255 (2.4)	236 (2.6)	220 (3.1)	204 (3.0)	188 (3.4)
Phosphorus, mg	1,421 (18.7)	1,363 (15.7)	1,310 (11.5)	1,253 (10.2)	1,201 (10.3)	1,153 (11.8)	1,104 (14.9)	1,055 (18.7)
Calcium, mg	963 (18.3)	927 (14.9)	896 (10.3)	861 (8.7)	830 (8.7)	801 (10.9)	772 (14.1)	741 (18.0)
Sodium, mg	3,570 (48.8)	3,465 (38.3)	3,370 (26.1)	3,266 (21.4)	3,172 (22.3)	3,084 (26.6)	2,993 (38.5)	2,900 (49.3)
Potassium, mg	2,975 (26.9)	2,787 (27.5)	2,613 (26.3)	2,437 (28.8)	2,275 (32.5)	2,123 (39.4)	1,974 (38.0)	1,830 (41.6)
Vitamin K, μg	89 (16.1)	82 (7.8)	76 (2.3)	69 (6.0)	64 (10.5)	59 (14.2)	54 (17.3)	50 (19.5)
Total Fiber, g	17 (0.3)	16 (0.2)	15 (0.2)	14 (0.2)	13 (0.2)	12 (0.2)	11 (0.2)	10 (0.2)
Total Choline, mg^b	350 (12.1)	330 (8.4)	308 (4.5)	289 (3.9)	270 (7.4)	254 (10.5)	236 (10.9)	220 (17.4)
^aBased on the Dietary Reference Intake (DRI) Life Stage Groups used in the Dietary Reference Intakes (Institute of Medicine (IOM), 1997) excluding children under 4 years, fasters, and pregnant and lactating women. Values were not controlled for total energy intake during the analyses.
^bTotal choline values available for NHANES 2005–2006 only and sample sizes for choline are shown in parentheses. Total choline content was calculated as the sum of free choline (Cho); lycerophosphocholine (GPC); phosphocholine (Pcho); phosphatidylcholine (PTdcho); and sphingomyelin (SM) (Patterson et al, 2008). Sample sizes for total choline were not adequate to generate percent greater than the AI.
^cThe NHANES provides values for folic acid and food folate in Dietary Folate Equivalents (DFE). We present folate as Dietary Folate Equivalents. IDFE = 1 μg food folate = 0.6 mg of folic acid from fortified food.

Appendix Table 5 Estimated Usual Intakes of Nutrients from Food [Median (SE)] by Range of Percent of Estimated Daily Intake from Added Sugars with Total Energy Intake as a Covariate in the Analysis, by Child Dietary Reference Intake (DRI) Life Stage Groups ^a for Vitamin A, Vitamin E, Vitamin C, Folate, Iron, Zinc, Magnesium, Phosphorus, Calcium, Sodium, Potassium, Vitamin K, Fiber and Total Choline ^b , National Health and Nutrition Examination Survey (NHANES) 2003–2006

	Categories of Individuals Based on Percent of Energy Intake from Added Sugar^c
	Children 4 to 8 Yrs			Males 9 to 13 Yrs			Females 9 to 13 Yrs			Males 14 to 18 Yrs			Females 14 to 18 Yrs
Nutrient	0 ⩽ x ⩽ 15%	15 < x ⩽ 25%	>25%	0 ⩽ x ⩽ 15%	15 < x⩽ 25%	>25%	0 ⩽ x ⩽ 15%	15 < x⩽ 25%	>25%	0 ⩽ x ⩽ 15%	15 < x ⩽ 25%	>25%	0 ⩽ x ⩽ 15%	15 < x ⩽ 25%	>25%0
n (Total Choline n)^b	840 (421)	694 (333)	146 (54)	433 (212)	450 (228)	126 (44)	477 (250)	445 (203)	117 (44)	470 (222)	620 (266)	261 (90)	449 (239)	568 (257)	233 (105)
Median (SE)
Vitamin A, μg RAE	517 (13.3)	423 (18.9)	352 (17.7)	576 (14.4)	466 (19.1)	358 (15.7)	547 (25.8)	464 (33.7)	351 (33.8)	695 (30.2)	545 (26.1)	421 (23.2)	544 (23.3)	479 (22.8)	384 (14.3)
Percent<EAR	(275μg/d)^d10%	20%	33%	(445μg/d) 31%	46%	66%	(420 μg/d) 29%	43%	63%	(630 μg/d) 43%	61%	73%	(485μg/d) 41%	51%	67%
Vitamin E, mg α-tocopherol	5.3 (0.17)	4.4 (0.11)	3.7 (0.25)	6.2 (0.33)	5.2 (0.15)	4.1 (0.18)	5.7 (0.18)	4.9 (0.28)	3.8 (0.31)	8.0 (0.32)	6.4 (0.24)	4.9 (0.32)	5.7 (0.14)	5.3 (0.17)	4.3 (0.15)
Percent<EAR	(6 mg/d) 63%	79%	84%	(9 mg/d) 80%	90%	96%	(9 mg/d) 87%	91%	96%	(12 mg/d) 76%	89%	87%	(12 mg/d) 96%	96%	98%
Vitamin C, mg	73 (2.9)	58 (1.2)	48 (2.4)	80 (4.9)	63 (1.6)	47 (2.1)	77 (2.1)	63 (6.5)	48 (2.9)	95 (3.4)	73(2.0)	56 (3.9)	76 (2.3)	65 (1.9)	52 (1.7)
Percent<EAR	(22 mg/d) 4%	8%	14%	(45 mg/d) 13%	24%	39%	(39 mg/d) 14%	23%	39%	(63 mg/d) 27%	41%	57%	(56 mg/d) 31%	40%	55%
Folate, μg^e	368 (7.4)	308 (7.1)	261 (12.0)	423 (16.1)	352 (9.2)	280 (11.9)	365 (10.6)	314 (17.5)	244 (17.3)	505 (16.4)	409 (14.5)	318 (15.7)	356 (8.4)	322 (9.7)	263 (7.1)
Percent<EAR	(160 μg/d) 1%	3%	10%	(250 μg/d) 7%	16%	37%	(250 μg/d) 13%	27%	52%	(330 μg/d) 16%	30%	53%	(330 μg/d) 42%	53%	72%
Iron, mg	13.8 (0.34)	12.3 (0.22)	11.2 (0.51)	16.3 (0.73)	14.4 (0.34)	12.3 (0.45)	13.5 (0.36)	12.4 (0.70)	10.3 (0.75)	19.6 (0.64)	16.8 (0.57)	13.9 (0.68)	13.2 (0.33)	12.8 (0.35)	11.1 (0.27)
Percent<EAR	(4.1 mg/d) 0%	0%	0%	(5.9m/d) 0%	1%	2%	(5.7 mg/d) 0%	2%	6%	(7.7mg/d) 1%	3%	9%	(7.9mg/d) 6%	9%	16%
Zinc, mg	10.4 (0.22)	9.0 (0.24)	8.0 (0.41)	12.5 (0.54)	10.9 (0.30)	9.0 (0.31)	10.1 (0.29)	9.0 (0.54)	7.3 (0.48)	15.3 (0.52)	12.9 (0.45)	10.2 (0.59)	9.8 (0.35)	9.3 (0.30)	7.9 (0.23)
Percent<EAR	(4.0 mg/d) 0%	0%	1%	(7mg/d) 4%	9%	23%	(7 mg/d) 12%	23%	46%	(8.5mg/d) 8%	15%	35%	(7.3m/d) 18%	23%	40%
Magnesium, mg	223 (3.6)	183 (4.2)	150 (7.9)	268 (11.1)	220 (5.5)	171 (6.1)	232 (6.3)	195 (10.4)	147 (10.6)	332 (10.4)	268 (8.6)	203 (12.1)	234 (5.5)	210 (6.3)	168 (4.9)
Percent<EAR	(110 mg/d) 1%	6%	21%	(200 mg/d) 20%	38%	67%	(200 mg/d) 32%	53%	79%	(340 mg/d)52%	73%	81%	(300 mg/d) 77%	85%	93%
Phosphorus, mg	1,206 (20.5)	1,039 (33.6)	908 (53.8)	1,407 (43.4)	1,230 (33.3)	1,011 (36.4)	1,189 (40.7)	1,055 (65.5)	830 (60.0)	1,682 (55.9)	1,447 (46.1)	1,150 (77.9)	1,147 (32.5)	1,092 (36.3)	919 (36.0)
Percent<EAR	(405 mg/d) 0%	0%	1%	(1055 mg/d) 19%	31%	55%	(1055 mg/d) 36%	50%	73%	(1055 mg/d) 12%	21%	43%	(1055 mg/d) 40%	46%	65%
Calcium, mg	944 (18.5)	817 (24.4)	716 (42.6)	1,029 (26.2)	884 (28.3)	725 (31.9)	928 (35.9)	828 (44.9)	660 (40.4)	1,200 (46.9)	997 (36.7)	798 (50.4)	869 (27.7)	817 (22.7)	688 (28.3)
Percent⩾AI	(800 mg/d) 67%	52%	40%	(1300 mg/d) 28%	17%	8%	(1300 mg/d)19%	14%	6%	(1300 mg/d) 43%	28%	21%	(1300 mg/d) 15%	13%	7%
Sodium, g	3,009 (50.9)	2,694 (60.8)	2,456 (158.8)	3,525 (125.1)	3,183 (82.7)	2,728 (103.8)	2,977 (84.7)	2,745 (144.3)	2,269 (144.9)	4,233 (139.5)	3,750 (114.4)	3,100 (211.5)	2,891 (73.3)	2,853 (76.8)	2,510 (112.4)
Percent⩾AI	(1.2 g/d) 100%	100%	98%	(1.5 g/d) 100%	98%	95%	(1.5 g/d) 98%	95%	86%	(1.5 g/d) 100%	99%	95%	(1.5 g/d) 97%	97%	92%
Potassium, g	2,144 (43.4)	1,762 (45.4)	1,446 (73.3)	2,570 (99.1)	2,134 (58.3)	1,662 (59.5)	2,210 (61.1)	1,870 (110.8)	1,412 (107.2)	3,142 (91.2)	2,578 (90.4)	1,973 (107.0)	2,237 (50.8)	2,017 (65.9)	1,614 (39.4)
Percent⩾AI	(3.8 g/d) 2%	1%	1%	(4.5 g/d) 6%	3%	0%	(4.5 g/d) 2%	1%	0%	(4.7 g/d) 17%	6%	9%	(4.7 g/d) 1%	2%	0%
Vitamin K, μg	49 (1.3)	38 (1.1)	30 (2.2)	58 (2.8)	44 (1.2)	32 (1.6)	60 (2.4)	48 (2.2)	34 (1.9)	80 (3.0)	58 (1.8)	42 (3.6)	64 (1.9)	55 (2.2)	41 (2.6)
Percent⩾AI	(55 μg/d) 41%	32%	12%	(60 μg/d) 47%	28%	12%	(60 μg/d) 49%	33%	14%	(75 μg/d) 54%	34%	22%	(75 μg/d) 38%	27%	14%
Total Fiber, g	12 (0.3)	9.9 (0.2)	8 (0.6)	14 (0.6)	12 (0.3)	9(0.4)	13 (0.4)	11 (0.6)	8 (0.5)	18 (0.6)	14 (0.5)	11 (0.7	13 (0.3)	12 (0.3)	9 (0.3)
Percent>AI (g/d)	(25 g/d) 2%	1%	1%	(31 g/d)4%	1%	0%	(26 g/d) 3%	2%	1%	(38 g/d) 7%	2%	4%	(26 g/d) 3%	3%	1%
Total Choline, mg^b	243 (5.6)	201 (5.6)	150 (19.5)	288 (9.0)	251 (10.3)	222 (52.0)	247 (9.9)	207 (10.1)	164 (14.4)	373 (48.1)	310 (21.7)	252 (44.3)	240 (6.5)	229 (7.7)	182 (17.4)
^aBased on the Dietary Reference Intake Life Stage Groups used in the Dietary Reference Intakes (Institute of Medicine (IOM), 1997) excluding children under 4 years, fasters, and pregnant and lactating women. Values were controlled for total energy intake during the analyses.
^bTotal choline values available for NHANES 2005–2006 only and sample sizes for choline are shown in parentheses. Total choline content was calculated as the sum of free choline (Cho); lycerophosphocholine (GPC); phosphocholine (Pcho); phosphatidylcholine (PTdcho); and sphingomyelin (SM) (Patterson et al., 2008). Sample sizes for total choline were not adequate to generate percent greater than the AI.
^cOur analyses employed two days of 24-hour dietary recall data from the National Health and Nutrition Examination Survey (NHANES), What We Eat in America 2003–2006. We restricted our sample to children and adults as defined by the DRIs as aged 4 years or older. Our sample was further restricted to individuals with reliable recall status, excluding fasters, pregnant women, and lactating women. The final analytic sample included 29,099 days of recall data from 15,189 individual respondents. For 2,539 individuals in the sample, only one 24-hour recall was reported.
^dEAR or AI for the respective life stage group.
^eThe NHANES provides values for folic acid and food folate in Dietary Folate Equivalents (DFE). We present folate as Dietary Folate Equivalents. IDFE = 1 μg food folate = 0.6 mg of folic acid from fortified food.

Appendix Table 6 Estimated Usual Intakes of Nutrients from Food [Median (SE)] by Range of Percent of Estimated Daily Intake from Added Sugars with Total energy Intake as a Covariate in the Analysis, by Adult Dietary Reference Intake (DRI) Life Stage Groups ^a for Vitamin A, Vitamin E, Vitamin C, Folate, Iron, Zinc, Magnesium, Phosphorus, Calcium, Sodium, Potassium, Vitamin K, Fiber and Total Choline ^b , National Health and Nutrition Examination Survey (NHANES) 2003–2006.

	Categories of Individuals Based on Percent of Energy Intake from Added Sugar^c
	Males 19 to 30 Yrs			Females 19 to 30 Yrs			Males 31 to 50 Yrs			Females 31 to 50 Yrs			Males 51 to 70 Yrs			Females 51 to 70 Yrs			Males 71+ Yrs			Females 71+ Yrs
Nutrient	0 ⩽ x⩽ 15%	15 < x⩽25%	>25%	0 ⩽ x⩽ 15%	15 < x⩽ 25%	>25%	0 ⩽ x⩽ 15%	15 < x⩽ 25%	>25%	0 ⩽ x⩽ 15%	15 < x⩽ 25%	>25%	0 ⩽ x⩽ 15%	15 < x⩽ 25%	>25%	0 ⩽ x⩽ 15%	15 < x⩽ 25%	>25%	0 ⩽ x⩽ 15%	15 < x⩽ 25%	>25%	0 ⩽ x⩽ 15%	15 < x⩽ 25%	>25%
n (Total Choline n)^b	495(201)	398(173)	204(81)	421(211)	323(143)	170(65)	807(377)	461(221)	171(74)	781(396)	383(177)	186(70)	817(368)	296(140)	102(51)	839(397)	298(130)	114(52)	562(238)	210(83)	36(13)	539(210)	220(78)	28(8)
Median(SE)
Vitamin A, μg RAE	686	586	498	559	487	379	703	593	460	560	474	366	597	519	393	526	439	387	541	417	328	506	419	335
	(16.2)	(15.8)	(31.1)	(15.4)	(16.5)	(25.9)	(12.3)	(35.8)	(33.9)	(23.4)	(41.2)	(17.7)	(17.1)	(23.0)	(29.1)	(14.5)	(21.8)	(17.6)	(11.7)	(19.7)	(23.8)	(20.8)	(31.1)	(18.5)
Percent<EAR	625 mg/d43%	55%	65%	500 mg/d 41%	52%	70%	625 mg/d 41%	54%	70%	500 mg/d 41%	54%	71%	625 mg/d 54%	65%	79%	500 mg/d 46%	61%	71%	625 mg/d 62%	79%	91%	500 mg/d 49%	65%	78%
Vitamin E, mg α-tocopherol	8.1	7.1	6.3	6.1	5.5	4.3	8.5	7.5	5.8	6.2	5.4	4.3	7.0	6.4	4.7	5.8	5.0	4.6	6.4	5.0	4.0	5.7	4.8	4.0
	(0.26)	(0.31)	(0.40)	(0.19)	(0.19)	(0.27)	(0.21)	(0.42)	(0.38)	(0.17)	(0.48)	(0.18)	(0.16)	(0.26)	(0.36)	(0.12)	(0.16)	(0.18)	(0.21)	(0.23)	(0.41)	(0.15)	(0.31)	(0.29)
Percent<EAR	12 mg/d78%	83%	88%	12 mg/d 93%	95%	97%	12 mg/d 77%	84%	89%	12 mg/d 94%	96%	98%	12 mg/d 88%	92%	92%	12 mg/d 97%	99%	99%	12 mg/d 95%	97%	100%	12 mg/d 98%	99%	99%
Vitamin C, mg	93	77	64	78	66	51	95	78	60	78	65	50	82	69	53	74	60	52	75	57	44	72	58	46
	(3.6)	(3.1)	(2.5)	(2.6)	(2.2)	(2.7)	(3.0)	(2.7)	(4.2)	(1.6)	(3.3)	(2.4)	(2.4)	(2.7)	(4.5)	(2.4)	(2.5)	(3.0)	(2.4)	(2.2)	(3.1)	(2.0)	(3.2)	(2.6)
Percent<EAR	75 mg/d36%	48%	59%	60 mg/d 34%	44%	60%	75 mg/d 36%	48%	63%	60 mg/d 33%	45%	61%	75 mg/d 44%	56%	70%	60 mg/d 37%	50%	60%	75 mg/d 50%	67%	82%	60 mg/d 38%	52%	66%
Folate, μg^e	495	430	380	361	322	255	500	435	342	356	308	244	417	373	282	331	282	253	375	294	235	312	264	215
	(10.4)	(10.9)	(21.4)	(8.3)	(9.0)	(14.7)	(7.5)	(27.6)	(23.5)	(10.6)	(25.3)	(9.8)	(10.1)	(14.9)	(14.8)	(6.1)	(10.7)	(9.0)	(6.1)	(13.9)	(22.7)	(9.6)	(17.1)	(11.9)
Percent<EAR	320 μg/d14%	24%	35%	320 μg/d 37%	49%	70%	320 μg/d 13%	23%	44%	320 μg/d 38%	54%	74%	320 μg/d 24%	34%	61%	320 μg/d 46%	65%	75%	320 μg/d 32%	59%	82%	320 μg/d 53%	72%	85%
Iron, mg	19.2	17.7	16.6	13.4	12.8	10.8	19.5	18.1	15.1	13.3	12.2	10.4	16.1	15.4	12.4	12.3	11.2	10.8	14.5	12.1	10.4	11.6	10.5	9.2
	(0.48)	(0.53)	(1.09)	(0.31)	(0.35)	(0.67)	(0.30)	(1.08)	(1.04)	(0.38)	(1.09)	(0.41)	(0.37)	(0.60)	(0.59)	(0.21)	(0.41)	(0.42)	(0.26)	(0.67)	(1.11)	(0.35)	(0.75)	(0.51)
Percent<EAR	6 mg/d0%	0%	1%	8.1 mg/d 8%	10%	23%	6 mg/d0%	0%	2%	8.1 mg/d 7%	13%	26%	6 mg/d0%	0%	2%	5 mg/d0%	0%	2%	6 mg/d0%	2%	3%	5 mg/d0%	1%	3%
Zinc, mg	15.1	13.6	12.5	10.0	9.4	7.6	15.4	13.9	11.2	9.9	9.0	7.4	12.5	11.7	9.0	9.2	8.1	7.7	11.2	9.0	7.6	8.6	7.6	6.4
	(0.37)	(0.43)	(0.81)	(0.25)	(0.27)	(0.53)	(0.32)	(0.93)	(0.84)	(0.29)	(0.72)	(0.31)	(0.29)	(0.48)	(0.49)	(0.13)	(0.25)	(0.31)	(0.31)	(0.51)	(0.74)	(0.23)	(0.52)	(0.35)
Percent<EAR	9.4 mg/d11%	18%	26%	6.8 mg/d 13%	18%	38%	9.4 mg/d 11%	18%	34%	6.8 mg/d 11%	22%	42%	9.4 mg/d 21%	26%	54%	6.8 mg/d 16%	26%	34%	9.4 mg/d 28%	55%	77%	6.8 mg/d 19%	35%	57%
Magnesium, mg	338	292	255	247	218	169	355	308	239	253	218	169	306	272	201	245	206	183	287	222	175	240	201	160
	(7.9)	(8.5)	(14.1)	(5.8)	(7.0)	(9.3)	(5.9)	(16.3)	(15.3)	(6.1)	(15.7)	(6.6)	(5.8)	(9.3)	(12.0)	(3.9)	(5.3)	(7.0)	(6.9)	(10.1)	(16.4)	(5.2)	(11.9)	(10.7)
Percent<EAR	330 mg/d48%	62%	72%	255 mg/d 53%	67%	83%	350 mg/d 48%	63%	80%	265 mg/d 56%	71%	86%	350 mg/d 86%	76%	86%	265 mg/d 60%	80%	87%	350 mg/d 75%	90%	99%	265 mg/d 64%	83%	90%
Phosphorus, mg	1,656	1,509	1,403	1,163	1,088	872	1,681	1,551	1,256	1,140	1,034	837	1,371	1,300	975	1,048	921	874	1,228	969	794	972	846	685
	(39.7)	(49.9)	(68.4)	(30.1)	(43.1)	(57.8)	(26.4)	(80.7)	(65.4)	(36.3)	(80.2)	(40.4)	(26.4)	(43.4)	(83.2)	(16.3)	(21.7)	(33.3)	(40.5)	(55.4)	(58.2)	(24.2)	(49.6)	(61.4)
Percent<EAR	580 mg/d0%	1%	3%	580 mg/d 3%	5%	18%	580 mg/d 0%	1%	6%	580 mg/d 2%	8%	22%	580 mg/d 1%	1%	7%	580 mg/d 3%	7%	13%	580 mg/d 1%	6%	17%	580 mg/d 4%	13%	32%
Calcium, mg	1,133	1,018	928	850	785	640	1,105	992	800	806	719	581	872	804	619	711	623	580	742	587	477	639	552	459
	(29.7)	(33.5)	(37.7)	(18.7)	(23.8)	(29.0)	(18.8)	(57.2)	(38.6)	(27.2)	(52.1)	(27.3)	(24.0)	(31.8)	(56.3)	(12.7)	(17.7)	(25.3)	(13.8)	(28.1)	(36.6)	(20.5)	(30.2)	(41.0)
Percent⩾AI	1000 mg/d61%	52%	44%	1000 mg/d 35%	28%	18%	1000 mg/d 59%	49%	33%	1000 mg/d 29%	23%	13%	1200 mg/d 23%	17%	13%	1200 mg/d 9%	5%	4%	1200 mg/d 11%	6%	1%	1200 mg/d 4%	2%	2%
Sodium, mg	4,184	3,926	3,773	2,946	2,852	2,400	4,264	4,047	3,401	2,903	2,734	2,323	3,495	3,420	2,695	2,686	2,459	2,430	3,149	2,599	2,244	2,511	2,285	1,961
	(104.2)	(125.7)	(183.6)	(94.9)	(118.9)	(141.8)	(70.4)	(227.8)	(181.6)	(75.5)	(185.7)	(103.4)	(76.5)	(118.7)	(203.8)	(40.0)	(50.9)	(83.1)	(92.9)	(141.8)	(144.2)	(46.1)	(117.9)	(164.4)
Percent⩾AI	1.5 g/d100%	99%	98%	1.5 g/d97%	94%	88%	1.5 g/d100%	99%	96%	1.5 g/d98%	94%	85%	1.3 g/d100%	100%	99%	1.3 g/d99%	98%	95%	1.2 g/d 100%	99%	98%	1.2 g/d 100%	98%	95%
Potassium, mg	3,203	2,792	2,471	2,365	2,098	1,623	3,354	2,951	2,327	2,425	2,098	1,634	2,930	2,634	1,971	2,349	1,996	1,780	2,766	2,173	1,732	2,311	1,945	1,552
	(66.8)	(70.0)	(148.8)	(52.3)	(56.1)	(119.8)	(44.1)	(160.6)	(148.2)	(57.5)	(157.2)	(69.9)	(55.4)	(90.8)	(92.6)	(34.3)	(56.6)	(64.4)	(61.9)	(107.1)	(164.7)	(50.7)	(127.0)	(89.9)
Percent⩾AI	4.7 g/d15%	11%	7%	4.7 g/d3%	2%	1%	4.7 g/d16%	9%	7%	4.7 g/d2%	1%	1%	4.7 g/d7%	4%	5%	4.7 g/d1%	0%	0%	4.7 g/d2%	2%	0%	4.7 g/d1%	0%	0%
Vitamin K, μg	86	69	56	72	60	44	97	77	55	79	63	46	86	71	50	80	63	54	83	60	45	83	65	50
	(3.9)	(4.4)	(2.1)	(2.6)	(3.0)	(1.9)	(4.5)	(2.4)	(3.2)	(2.2)	(3.0)	(2.5)	(3.0)	(3.4)	(6.4)	(3.3)	(2.2)	(3.9)	(4.4)	(2.2)	(2.6)	(1.8)	(1.9)	(6.4)
Percent⩾AI	120 μg/d30%	20%	13%	90 μg/d 34%	23%	11%	120 μg/d 36%	23%	12%	90 μg/d 39%	26%	12%	120 μg/d 26%	17%	10%	90 μg/d 40%	23%	15%	120 μg/d 23%	10%	2%	90 μg/d 43%	24%	13%
Total Fiber, g	18	16	14	14	12	9	19	16	13	14	12	9	16	14	11	14	12	10	15	12	9	13	11	9
	(0.4)	(0.5)	(0.6)	(0.3)	(0.4)	(0.5)	(0.3)	(0.9)	(0.9)	(0.4)	(1.0)	(0.4)	(0.4)	(0.6)	(0.7)	(0.3)	(0.5)	(0.4)	(0.3)	(0.7)	(0.9)	(0.5)	(0.8)	(0.4)
Percent⩾AI	38 g/d6%	3%	3%	25 g/d7%	4%	2%	38 g/d5%	3%	2%	25 g/d6%	3%	1%	30 g/d6%	3%	4%	21 g/d11%	4%	3%	30 g/d2%	1%	0%	21 g/d8%	3%	2%
Total Choline, mg^b	390	344	295	259	222	174	417	368	308	276	231	182	379	335	267	263	230	207	348	269	253	260	224	175
	(26.3)	(18.8)	(12.1)	(12.1)	(9.7)	(12.9)	(10.2)	(23.4)	(19.0)	(6.5)	(6.5)	(10.1)	(7.3)	(13.6)	(51.1)	(4.3)	(6.2)	(12.0)	(7.9)	(7.6)	(50.6)	(5.4)	(9.3)	(26.0)
^aBased on the Dietary Reference Intake Life Stage Groups used in the Dietary Reference Intakes (Institute of Medicine (IOM), 1997) excluding children under 4 years, fasters, and pregnant and lactating women. Values were controlled for total energy intake during the analyses.
^bTotal choline values available for NHANES 2005–2006 only and sample sizes for choline are shown in parentheses. Total choline content was calculated as the sum of free choline (Cho); lycerophosphocholine (GPC); phosphocholine (Pcho); phosphatidylcholine (PTdcho); and sphingomyelin (SM) (Patterson et al, 2008). Sample sizes for total choline were not adequate to generate percent greater than the AI.
^c Our analyses employed two days of 24-hour dietary recall data from the National Health and Nutrition Examination Survey (NHANES), What We Eat in America 2003–2006. We restricted our sample to children and adults as defined by the DRIs as aged 4 years or older. Our sample was further restricted to individuals with reliable recall status, excluding fasters, pregnant women, and lactating women. The final analytic sample included 29,099 days of recall data from 15,189 individual respondents. For 2,539 individuals in the sample, only one 24-hour recall was reported.
^dEAR or AI for the respective life stage group.
^eThe NHANES provides values for folic acid and food folate in Dietary Folate Equivalents (DFE). We present folate as Dietary Folate Equivalents. IDFE = 1 μg food folate = 0.6 mg of folic acid from fortified food.

Appendix Table 7 Mean Intake of Added Sugar in Gram-Equivalents (g-eq) ^a by Food Category for Dietary Reference Intake (DRI) Life Stage Groups ^b with Total Energy Intake as a Covariate and Covariates from the National Health and Nutrition Examination Survey (NHANES) Food Frequency Questionnaire in the Analysis, NHANES 2003–2006

	DRI Life Stage Groups
Food Category	Children4 to 8 yrs	Males9 to 13 yrs	Females9 to 13 yrs	Males14 to 18 yrs	Females14 to 18 yrs	Males19 to 30 yrs	Females19 to 30 yrs	Males31 to 50 yrs	Females31 to 50 yrs	Males51 to 70 yrs	Females51 to 70 yrs	Males71+ yrs	Females71+ yrs
n	1,136	688	749	836	901	578	641	954	987	858	956	594	554
Mean g-eg (SEM)^c
Total added sugar	92.2 (1.73)	103.4 (2.37)	87.2 (2.16)	115.2 (3.02)	81.7 (1.62)	117.7 (4.02)	76.3 (1.56)	106.3 (2.24)	69.2 (1.36)	78.2 (1.96)	58.0 (1.32)	58.9 (1.67)	47.3 (1.29)
Milk and milk products	11.0 (0.41)	10.1 (0.61)	10.9 (0.58)	10.7 (0.61)	8.6 (0.34)	9.6 (0.52)	8.1 (0.43)	8.4 (0.33)	7.7 (0.34)	6.9 (0.29)	6.8 (0.31)	5.9 (0.38)	6.1 (0.23)
Meat/poultry/fish/eggs	1.1 (0.05)	1.4 (0.06)	1.0 (0.05)	1.6 (0.08)	1.0 (0.04)	1.7 (0.08)	1.0 (0.04)	1.6 (0.07)	0.9 (0.04)	1.2 (0.06)	0.8 (0.03)	1.0 (0.06)	0.7 (0.03)
Legumes	0.3 (0.03)	0.4 (0.05)	0.3 (0.03)	0.6 (0.07)	0.3 (0.03)	0.7 (0.14)	0.4 (0.03)	0.8 (0.11)	0.4 (0.05)	0.7 (0.08)	0.5 (0.05)	0.7 (0.07)	0.5 (0.05)
Grain products ^d	21.1 (0.63)	22.8 (1.04)	19.6 (0.69)	25.5 (1.03)	18.3 (0.63)	24.0 (0.89)	17.1 (0.53)	22.0 (0.67)	15.9 (0.43)	18.2 (0.80)	14.6 (0.48)	17.0 (0.80)	14.9 (0.64)
Breakfast cereal	6.6 (0.36)	6.7 (0.40)	5.1 (0.37)	6.2 (0.27)	3.9 (0.25)	4.6 (0.25)	3.0 (0.15)	3.5 (0.21)	2.4 (0.13)	2.8 (0.18)	2.1 (0.13)	2.6 (0.20)	2.0 (0.14)
Sweetened grains	8.7 (0.31)	10.2 (0.52)	9.5 (0.42)	14.2 (0.93)	9.7 (0.38)	14.8 (0.91)	9.7 (0.40)	14.2 (0.54)	9.5 (0.34)	10.9 (0.63)	8.7 (0.37)	9.3 (0.48)	8.6 (0.45)
Other grains	4.3 (0.15)	4.7 (0.17)	4.1 (0.17)	5.3 (0.19)	4.0 (0.13)	5.5 (0.17)	3.9 (0.10)	5.4 (0.16)	3.8 (0.10)	4.6 (0.16)	3.6 (0.08)	4.1 (0.17)	3.5 (0.12)
Fruits	0.6 (0.11)	0.7 (0.11)	0.7 (0.12)	0.7 (0.09)	0.7 (0.08)	0.8 (0.10)	0.7 (0.08)	0.8 (0.09)	0.7 (0.07)	0.8 (0.10)	0.9 (0.08)	0.9 (0.11)	1.0 (0.10)
Vegetables	0.9 (0.06)	1.1 (0.08)	0.8 (0.16)	1.2 (0.09)	0.9 (0.07)	1.3 (0.13)	0.8 (0.05)	1.2 (0.08)	0.7 (0.05)	1.0 (0.07)	0.7 (0.05)	0.9 (0.06)	0.7 (0.06)
Fats/oils/salad dressings	0.2 (0.03)	0.3 (0.03)	0.3 (0.03)	0.4 (0.04)	0.4 (0.04)	0.5 (0.05)	0.5 (0.04)	0.6 (0.05)	0.7 (0.06)	0.7 (0.05)	0.8 (0.06)	0.7 (0.06)	0.8 (0.07)
Sugars/sweets/beverages ^e	51.5 (33.94)	65.8 (13.81)	51.2 (5.51)	83.4 (2748)	58.6 (33.52	84.3 (31.11)	51.6 (23.51)	69.6 (12.08)	40.5 (4.37)	44.2 (12.78)	28.9 (8.69)	30.9 (13.39)	22.7 (5.60)
Sugars/sweets	12.2 (0.53)	13.6 (0.64)	12.2 (0.51)	14.7 (0.62)	11.7 (0.45)	14.7 (0.80)	10.9 (0.36)	13.7 (0.63)	10.3 (0.32)	11.2 (0.59)	9.1 (0.32)	9.9 (0.48)	8.3 (0.25)
Regular soft drinks	21.1 (1.09)	31.0 (1.54)	22.3 (1.61)	44.6 (1.99)	28.8 (1.84)	47.6 (2.40)	25.6 (1.66)	39.2 (1.79)	19.3 (1.32)	22.8 (0.96)	13.0 (0.80)	15.1 (0.86)	9.4 (0.60)
Regular fruitades/drinks	15.7 (0.64)	17.5 (1.18)	12.2 (0.60)	17.1 (1.02)	11.1 (0.63)	13.5 (1.07)	8.2 (0.67)	9.5 (0.63)	5.8 (0.38)	5.6 (0.42)	3.7 (0.29)	3.5 (0.32)	2.5 (0.21)
Low calorie beverages	0.0 (0.00)	0.0 (0.00)	0.0 (0.00)	0.0 (0.00)	0.0 (0.00)	0.0 (0.00)	0.0 (0.00)	0.0 (0.01)	0.0 (0.01)	0.0 (0.01)	0.0 (0.01)	0.0 (0.01)	0.0 (0.01)
Alcoholic beverages	0.1 (0.04)	0.1 (0.05)	0.2 (0.08)	0.4 (0.13)	0.4 (0.16)	0.9 (0.23)	0.4 (0.10)	0.8 (0.17)	0.5 (0.10)	0.4 (0.08)	0.5 (0.10)	0.3 (0.09)	0.5 (0.12)
Other non-alcoholic beverages	2.2 (0.17)	2.8 (0.22)	2.4 (0.21)	3.6 (0.32)	2.7 (0.25)	4.5 (0.63)	3.2 (0.34)	4.9 (0.52)	3.7 (0.37)	4.7 (0.52)	3.2 (0.27)	3.7 (0.37)	2.7 (0.27)
^aOne Gram–equivalent equals an amount of added sugar comparable to 1 g sucrose in carbohydrate content.
^bBased on the Dietary Reference Intake Life Stage Groups used in the Dietary Reference Intakes (Institute of Medicine (IOM), 1997) excluding children under 4 years, fasters, and pregnant and lactating women. Values were controlled for total energy intake during this analyses.
^cThe standard errors reported in this table have been adjusted via balanced repeated replication, and therefore account for the complex NHANES sampling design.
^dExamples of grain products sub-categories include: breakfast cereals: sweetened and unsweetened cereals, cooked cereals, rice cereals, cereals not cooked or unknown as to cooked; sweetened grains: cakes, cookies, pies, pastries, crackers and salty snacks from grains, funnel cakes; other grains: four and dry mixes, muffins, yeast breads, yeast rolls, quick breads, pancakes, waffles, French toast, crepes, four-water patties, flour-milk patties, rice flour cakes, grain mixtures, meat substitutes that are mail cereal protein.
^eExamples of sugars/sweets/beverages sub-categories include: sugars/sweets: sugar, sugar substitutes, candy, ice cream toppings, syrups, chewing gum, gum drops, ice pops, gelatin desserts, jams, jellies; regular soft drinks: cola and non-cola type, cream sodas. Ginger ales, root beer, chocolate-flavored soda; regular fruitades/fruit drinks: all single fruit and mixed fruit beverages such as apple juice, apple-cranberry-grape juice, fruit punch, lemonade, fruit flavored sports drinks; fruit flavored powdered drinks; low calorie beverages: any sugar-free or low calories beverage, carbonated or not carbonated; alcoholic beverages: any alcoholic drinks including beer, wine, cocktails; other non-alcoholic beverages: coffee, tea, coffee substitutes, non-fruit beverages, non-alcoholic beer, wines, cocktails, non-carbonated water.

ACKNOWLEDGEMENTS

We thank Fran Seligson, Ph.D., and Patty Connor for helpful suggestions during manuscript revisions; the staff at NCI and in the Food Surveys Research Group, ARS, USDA for helpful discussions and recommendations about using the NCI usual intake method; Lois Steinfeldt, for feedback on data collection procedures and data preparation; Bill Rhodes for consultation on methodology; and Janice Abbas for manuscript preparation. We appreciate the support and organization that Marie Latulippe of ILSI, NA brought to the project.

Funding was provided by the Technical Committee on Carbohydrates of the International Life Sciences Institute, North American Branch.

Author disclosures: B. P. Marriott, L. Olsho, L. Hadden and P. Connor have no conflicts of interest.

Notes

For 2,539 individuals in the sample, only one 24-hour recall was reported.

The full detailed results for the 13 DRI life stage groups defined by percent energy intake from added sugars for eight groups in five-percentage-point increments, 0–5% through 35%+ are available from the first author.

An alternative approach to taking the simple average would have been to calculate the energy-weighted average percentage intake, thereby adjusting for within-person variation in energy intake by day. As a sensitivity check, we tested this alternative classification method, and found that less than 6% of the sample would be assigned to a different added sugars intake group using this alternative method. We felt this difference in classification was unlikely to greatly influence our results, and therefore we employed the simple average in our analyses.

Since the analyses for this paper were performed, NCI has released a new SAS macro (“INDIVINT”) that may be used to estimate usual intake at the individual level. Had this new methodology been available at the time we performed our analyses, we would have employed it in generating individual-level added sugars intake estimates for classification purposes. Readers seeking to replicate or expand on our work are encouraged to employ this methodology.

The MIXTRAN and DISTRIB macros (and the new INDIVINT macro described above) are available for free download at the NCI website: http://riskfactor.cancer.gov/diet/usualintakes/macros.html

Data on total choline intake were available only for NHANES 2005–2006.

Because data on total choline were available only in NHANES 2005–2006, sample sizes were too small to generate comparisons to the AI level for this nutrient.

Data not shown and available from the first author upon request.

Comparable tables including the data for the thirteen nutrients plus fiber for the thirteen DRI life stage groups divided into the eight added sugars groups from 0 to > 35% intake of energy can be obtained by contacting the first author.

Note the numbers may not total exactly due to rounding.

The USDA report cited here with the usual intake estimates using the ISU method can be obtained from: http://www.ars.usda.gov/research/publications/Publications.htm?seq_no_115=184176

It should be noted that results obtained via the NCI method might be expected to differ more greatly from estimates obtained via the ISU method for episodically-consumed foods than for the type of nutrient estimates presented here, since the amount-only specification we apply for nutrients does not take advantage of the NCI method's capability to account for correlation between probability and amount of consumption.