Sugars are a major contributor of energy in the US diet.1 These include added sugars (ASs), caloric sweeteners that are added to foods during processing or preparation, and naturally occurring sugars (NOSs), those that are an integral part of many commonly consumed foods (eg, dairy products, fruits, and vegetables). Added sugars are chemically indistinguishable from NOSs but tend to be found in foods that provide little or no additional nutrients.1 The consumption of foods high in NOS is generally associated with positive diet and health outcomes, whereas the consumption of foods high in AS, particularly in the form of sugar-sweetened beverages (SSBs), has been associated with decreased diet quality,2 increased risk of dental caries,3 increased energy intake and risk of obesity,4–6 and increased risk of obesity-associated conditions, including diabetes,7,8 cardiovascular disease,9–11 and fatty liver disease.12
The most common AS in the US food supply are sucrose (ie, table sugar) and high-fructose corn syrup.13 Sucrose is a disaccharide composed of the monosaccharides fructose and glucose. Similarly, high-fructose corn syrup is made from approximately equal parts of fructose and glucose. Although there is conflicting evidence, several mechanisms have been proposed to explain the association between AS intake and its associated outcomes, some of which relate to the glucose component, some relate to the fructose, and others relate to the liquid form in which much of ASs are consumed, namely, in sodas and other SSBs. A high intake of glucose results in a corresponding release of insulin. This insulin release drives down blood glucose levels, which may promote feelings of hunger and increased calorie consumption in the short term and decreased insulin sensitivity,14 particularly among those who are overweight or obese,7 during the long term. Unlike glucose, the metabolism of fructose takes place almost exclusively in the liver and does not elicit an insulin response. Studies suggest that, when fructose is consumed in large quantities, there is a rapid influx into the liver that can lead to an increase in de novo lipogenesis and elevated triglyceride levels,15–17 raising the risk of cardiovascular and liver diseases. There is controversy as to whether the observed effects are due to fructose itself or to the results of an overconsumption of energy.18 It has also been suggested that, when sugars are consumed in liquid form, as in SSBs and in fruit juices, they may not be compensated for in the same way, or to a lesser degree, as sugars consumed in foods.18–20 One hypothesis is that the impact of liquid forms of AS on satiety contributes to overconsumption and an increased risk of obesity and associated chronic conditions. Although it has been proposed that this is the result of liquid’s limited impact on satiety, this remains controversial.21
An analysis of the dietary data from the 2007 to 2008 National Health and Nutrition Examination Survey (NHANES) demonstrated that, among individuals 2 years or older, AS contributed, on average, 14.6% of the total energy consumption, with consumption highest among African Americans and those in the lowest-income quartile.22 Very young children are known to have a preference for sweet taste that may influence long-term dietary patterns,23 but there has been little study of the role of sugars in their diets. Results from the 2002 Feeding Infants and Toddlers Study demonstrated that, by the age of 19 to 24 months, 62% of children consumed baked desserts, 20% consumed candy, and 44% consumed SSBs.24 Sugar-sweetened beverages were the third leading source of energy for toddlers aged 12 to 24 months, with milk and 100% fruit juices ranking first and second, respectively.25 Currently, the 2015 to 2020 Dietary Guidelines for Americans recommend that AS intake be limited to less than 10% of the total energy intake to facilitate meeting nutrient intake requirements without exceeding total calorie needs.26
The purposes of this study were to estimate the current intake of AS among US toddlers aged 12 up to 24 months and determine whether consumption patterns differ by sex, race/ethnicity, and household income. We also aimed to identify the leading food and beverage sources of AS among toddlers. Results of this study will inform the development of dietary guidelines, as well as efforts by the food industry to provide products that help meet young children’s nutrient needs without compromising their immediate or long-term diet and health.
Data from NHANES were used for this descriptive analysis. The NHANES is a continuous cross-sectional survey designed to monitor health and nutrition among individuals in the United States. Demographic data—age, race/ethnicity, and household poverty income ratio—and dietary data were collected from a parent or guardian of toddlers aged 12 up to 24 months (n = 479) enrolled in the 2009 to 2010 and 2011 to 2012 cycles of NHANES. For the purposes of examining trends from infancy, AS intake among nonbreastfed infants aged 0 to 11 months (n = 248) was also assessed. A detailed description of NHANES data collection and sampling procedures can be found elsewhere.27 The institutional review board of the National Center for Health Statistics approved the NHANES data collection protocol.
Dietary data were collected using 1 interviewer-assisted, in-person 24-hour dietary recall. The NHANES uses a multipass, computer-assisted process to collect dietary data. An estimate of total sugar intake, calculated as the sum of all grams of sugars consumed by each participant, was provided as a variable on the NHANES dietary data sets for each cycle. For this analysis, intakes of 4 different sugar types and forms were estimated: (1) AS in beverages (SSBs, including sodas, sweetened fruit drinks, sweetened teas, coffees, flavored milks, and sports drinks), (2) AS in foods (cakes, pies, candies, etc), (3) NOS in beverages (100% fruit juices and milks), and (4) NOS in foods (fruits, vegetables, and dairy products). To identify the amount of each type and form of sugars consumed, individual food codes and descriptions were used to categorize each food and beverage item. Because the NHANES dietary data sets do not contain information on AS content, data from the 2009 to 2010 and 2011 to 2012 NHANES cycles were merged with the AS content information provided in the Food Patterns Equivalents Database (FPED) from 2009 to 2010 and 2011 to 2012, which contains an estimate of the AS content of all foods consumed by NHANES study participants.28 For each food and beverage item, the amount of NOS was estimated by subtracting the amount of AS from the amount of total sugars.
The amounts (in grams) of each type and form of sugar contained in all foods and beverages consumed were summed to estimate the total intake for each individual during the previous 24 hours. These total intake values were multiplied by 4 kcal/g to estimate the amount of energy provided by each. To determine the percentage of total energy (% total energy) contributed by each sugar type and form, these values were then divided by each individual’s total calorie intake for the day. To identify the major sources of ASs, foods and beverages were categorized based on the food and beverage coding scheme used in the USDA Food and Nutrient Database for Dietary Studies.29,30 The food codes and descriptions of the foods and beverages included in each category can be found in a supplemental Table (Supplemental Digital Content 1, http://links.lww.com/NT/A18). To account for sample weighting, percentages of total AS consumed and the rank order by food category were calculated from the population average rather than by summing all ASs for each category and dividing by the total amount of AS consumed by our sample. Intake was assessed for all children by demographic subgroups and by occasion and location of consumption.
Two-sample t tests and pairwise comparisons using differences of least squares means estimates (via PROC SURVEYREG LSMESTIMATE statement) were conducted to compare mean intake (±standard errors) across demographic and AS intake subgroups. Tests for trend using linear orthogonal contrasts were conducted to assess differences in macronutrient intake and body weight with increasing AS intakes and to assess trends in AS consumption with age in months. Trends and comparisons by increasing AS intakes were adjusted for age, sex, race/ethnicity, and poverty-to-income ratio. SAS 9.4 software (Cary, North Carolina) was used for all analyses. Sampling weights were applied to obtain nationally representative estimates. Procedures to adjust the variances for the complex sampling methods used in NHANES were used for all analyses.
United States toddlers (12 up to 24 months old) consumed an average of 90.5 g/d of total sugars (Table 1), which represented 29.6% of their total energy intake (not shown). Most of the sugars consumed were NOS (64.4 ± 2.5 g), and approximately half of the NOSs were from dairy sources (29.6 ± 1.0 g). Mean daily intake of AS was 26.2 ± 1.3 g, which represented 30.1% of the total sugar and 8.4% of the total energy intake (Table 2). Of the AS consumed, two-thirds were consumed in foods (15.7 ± 0.8 g), and one-third were consumed in beverages (10.5 ± 1.1 g) (Table 1).
The consumption of AS begins early and rises, both in absolute terms and as a percentage of total energy intake, throughout the infancy and toddler periods (P-trend < .0001) (Figures 1, 2). The linear trend was unchanged when that analysis was adjusted for weight (kilograms), race/ethnicity, and household income. Infants younger than 12 months consumed an average of 2.8 g/d of AS. Among toddlers aged 12 up to 24 months, the consumption of AS was significantly higher (P < .05) among non-Hispanic black children (33.4 ± 2.7 g) than among non-Hispanic white or Mexican children (24.8 ± 2.1 and 25.6 ± 3.2 g, respectively) and lower among those from the highest-income households (23.2 ± 2.7 g) compared with those from middle- and lowest-income households (28.4 ± 3.1 and 26.3 ± 1.7 g, respectively). No significant difference in AS consumption was observed between male and female toddlers (Table 1).
The toddlers’ AS consumption was positively correlated with total carbohydrate and total sugar intake. As AS intake rose from less than 5% to 15% or greater of total energy, total carbohydrate intake rose from 51.0% to 59.5% of total energy (P-trend = .006), and total sugar intake rose from 28.4% to 38.7% of total energy (P-trend < .001) (Table 2). The toddlers’ AS consumption was inversely correlated with fiber, protein, and saturated fat intake. As AS intake rose from less than 5% to 15% or greater of total energy, daily intake of fiber decreased from 9.0 ± 0.6 to 7.5 ± 0.5 g (P-trend = .06), protein decreased from 16.4% to 13.2% of total energy intake (P-trend = 0.003), and saturated fat decreased from 14.4% to 10.6% of total energy intake (P = .001) (Table 2). Comparing macronutrient intake among the lowest AS consumer (<5% of total energy) with those in each of the higher categories of AS intake, carbohydrate and total sugar intakes were significantly higher, and total and saturated fat and protein intakes were significantly lower only among those consuming 15% or greater of their total energy as ASs.
Sweetened fruit juices and fruit flavored drinks were the largest contributors of AS (23.3%) in the toddlers’ diet (Table 3). Other top contributors to AS intake included cakes/cookies/pastries/pies (15.3%), sugars/sweets (10.3%), cereals/rice/pasta (8.4%), and yogurt (7.3%). In addition, milk/milk drinks contributed 6.9% of the AS consumed, and sodas/other sugar-sweetened soft drinks contributed 5.7% of the AS consumed. The AS in sweetened fruit juices/fruit flavored drinks contributed 2.0% of the toddlers’ total energy intake, whereas the AS in all SSBs combined (excluding milk) contributed 33.3% of all AS consumed and 2.7% of total energy intake (not shown).
Slightly more than half of the ASs were consumed as part of a meal (54.8%), with the remainder consumed between meals (as snacks). One-quarter of ASs are consumed at breakfast (morning meal), 17% at lunch (midday meal), and 13% at dinner (evening meal) (not shown). Most AS consumption occurred at home (83.1%), and 3.7% was consumed at fast food restaurants.
Sugars, both added and naturally occurring, are a major contributor of energy in the US diet. Given the link between AS consumption and increased risk of obesity and chronic disease, both public health experts and the popular press have focused attention on the high consumption levels. This study was the first to estimate total AS intake among children younger than 2 years, identify the leading sources of AS for this age group, and describe how this intake correlates with the intake of other macronutrients.
Our analysis found that toddlers consumed 8.4% of their total energy as ASs. Previous research among children 2 to 5 and 6 to 11 years old estimated that AS contributed 13.4% and 17.0%, respectively, of their total energy intake.19 Similarly, we demonstrated that AS in SSBs contributed 2.7% of the total energy consumed by children aged 12 up to 24 months, whereas previous research estimated that SSBs contributed 4.0% and 6.2% of the total energy consumed by slightly older children, those 2 to 5 and 6 to 11 years old, respectively.19 This suggests that total AS consumption and consumption of AS in beverages begin early and continue to rise throughout early childhood.
Eating behaviors evolve during the first years of life.31 The milk-based diet of infancy rapidly transitions to an increasingly diverse diet.24 Because infants are born with a preference for sweet and salty tastes, sweet and salty foods tend to be more accepted than mostly bitter foods, such as many vegetables.31,32 Research has shown that the transition into late childhood is characterized by increased consumption of unhealthy snacks and SSBs, which are major sources of AS.31 It has been hypothesized that a diet high in sweet-tasting foods in early childhood leads to a greater consumption of AS and its associated health outcomes long term. Sonneville et al33 recently demonstrated that higher juice intake at 1 year was associated with higher juice intake, higher SSB intake, and greater body mass index z score during early and middle childhood. A study using data from the longitudinal Infant Feeding Practices Study II showed that, by their first year, approximately two-thirds of infants consumed high-fat/sugar foods.34
The use of data from a large national sample made it possible for the first time to estimate the amount of AS being consumed by US children younger than 2 years. The use of the most current dietary data available provides a useful approximation of current intake, and the large sample size allows for the examination of intake by age and across demographic subgroups. The availability of data from 24-hour dietary recalls provides a means to examine the relationship between AS consumption levels and intakes of other nutrients.
Limitations of this study include the use of parent- or guardian-reported intakes. Little is known about the validity of these intake reports for young children, but there is a potential for information bias resulting from the underreporting of AS given current recommendations to minimize its intake23 and given that many children attend out-of-home child care, which limits the parents’ ability to know what they consumed during the day. Because of expected day-to-day variations, a single 24-hour dietary recall is unlikely to provide a valid estimate of an individual’s usual intake; however, data from one 24-hour dietary recall are sufficient to estimate mean usual intake for a group as was conducted in our analysis. A more accurate method for assessing children’s usual intake would provide a better estimate of the proportion of children meeting the recommended AS intake level. In addition, the sugar content of foods and beverages was not directly measured.35 This information has to be estimated using ingredient lists, food labels, and industry-provided data, when available. Finally, in the FPED 2011 to 2012, fruit juice concentrates added to foods and beverages were categorized as ASs, whereas these additions were categorized as fruit juices in the FPED 2009 to 2010. This suggests that the AS intake estimates reported for our 2009 to 2012 study period may be underestimated.
The consumption of AS begins in infancy and increases throughout the toddler period. By their second birthday, mean AS consumption approaches the recommended limit. Added sugars are consumed primarily in foods and beverages that tend to be low in other nutrients important in the diets of young children. Efforts are needed to ensure that AS intake is minimized among young children and that foods that contain them contribute meaningful amounts of those nutrients that tend to be lacking in their diets.
Supplemental Digital Content