Food Sources of Energy and Nutrients of Public Health Concern and Nutrients to Limit with a Focus on Milk and other Dairy Foods in Children 2 to 18 Years of Age: National Health and Nutrition Examination Survey, 2011–2014

Many children are not meeting current nutrient recommendations. The objective of this study was to determine the food sources of energy, nutrients of public health concern, and nutrients to limit with a focus on dairy foods. Twenty-four-hour dietary recall data from children 2–5 (n = 1511), 6–11 (n = 2193), and 12–18 years (n = 2172) participating in NHANES 2011–2014 were analyzed. Energy, fiber, calcium, potassium, vitamin D, added sugars, saturated fatty acids (SFA), and sodium intakes were sample-weighted and ranked on percentage contribution to the diet using specific food group intake and disaggregated data for dairy foods. For children 2–5, 6–11, and 12–18 years, milk, sweet bakery products, and sweetened beverages, respectively were the top food sources of energy, respectively. For calcium, potassium, and vitamin D, milk was the top ranked food source in all age groups. For children 2–5, 6–11, and 12–18 years, milk, sweet bakery products, and pizza, respectively were the top three ranked food sources of SFA; and sugar sweetened beverages and sweet bakery products were to top two food group sources of added sugars. Cured meats/poultry, pizza, and pizza, respectively, were the top ranked food sources of sodium for the three age groups. Identification of food sources of these nutrients can help health professionals implement appropriate dietary recommendations and plan age-appropriate interventions.


Introduction
Dairy products are rich in three of the four nutrients of public health concern: calcium, vitamin D, and potassium [1]. The 2015-2020 Dietary Guidelines Advisory Committee (DGAC) determined that several nutrients: vitamins A, E, and C; folate; magnesium; and iron (in adolescent females) were under consumed relative to the Estimated Average Requirement (EAR) or Adequate Intake (AI) levels set by the Institute of Medicine and these were characterized as "shortfall nutrients". The DGAC confirmed that fiber, calcium, vitamin D, and potassium remained nutrients of public health concern since underconsumption has been linked to adverse health outcomes [2]. In children 2-18 years of age (years), milk has previously been shown to be the primary source of calcium, vitamin D, and potassium [3]. Dairy products, especially milk and yogurt, also provide protein, saturated fatty acids (SFA), riboflavin, vitamin B 12 , and phosphorus.

Study Overview, Study Population, and Analytic Sample
The NHANES is a program of studies designed to assess the health and nutritional status of free-living individuals in the US. Online information about the NHANES, including the purpose [25], plan and operations, sampling and weighting procedures, analytic guidelines [26], response rates, and population totals [27], is available. Data from children 2-18 years of age (years) participating in the NHANES from 2011 to 2014 were used for these analyses. The final analytic sample had 5876 participants; children were separated into three age groups: 2-5 years (n = 1511), 6-11 years (n = 2193), and 12-18 years (n = 2172). The National Center for Health Statistics (NCHS) Research Ethics Review Board has approved the use of human subjects for NHANES studies [28]; and further institutional review was not required.

Dietary Intake
Dietary intake data for the NHANES used in this study were obtained from the in-person 24-h dietary recall interview [29] using an Automated Multiple-Pass Method [30]. Although a second, telephone interview, was also taken 3 to 10 days after the in-person interview, only the in-person interview was used because of the difference in the methodology for collecting the two recalls. A single 24-h dietary recall administered in a large population can provide data to adequately estimate population mean intakes [31]. Survey participants 12 years and older completed their own dietary interview; children 6 to 11 years were assisted by an adult, usually a parent; and parents/guardians reported for children younger than 6 years [29].

Food Groupings and Composition
The relevant What We Eat in America (WWEIA), the dietary component of NHANES, food category classification systems [32] were used to classify all foods. The WWEIA food categories contain 15 main groups: milk and dairy; protein foods; mixed dishes; grains; snacks and sweets; fruit; vegetables; beverages, nonalcoholic; alcoholic beverages; water; fats and oils; condiments and sauces; sugars; infant formula and baby food; and other. The WWEIA food categories also consists of 47 subgroups. For example, for the milk and dairy main group, the subgroups were milk, flavored milk, cheese, dairy drinks and substitutes, and yogurt. For these analyses we focused on the 47 subgroups.
Using the relevant Food Patterns Equivalent Database [33] milk, cheese, and yogurt servings of non-dairy foods and especially mixed dishes were determined. The nutrient composition in the relevant Food and Nutrient Database for Dietary Studies FNDDS 2011-2012 and 2013-2014 [34] linked to SR 26 and SR 28 respectively [1] for milk, NFS (not further specified); cheese, NFS; and yogurt, NFS was used to assess energy and nutrient contribution of dairy servings non-dairy foods. The nutrients reported herein are the nutrients of public health concern [2]: dietary fiber, calcium, vitamin D, and potassium; and nutrients to limit: SFA, added sugars, and sodium.
Data are reported as specific food group (SFG) intake, adjusted intake, and delta intake. Specific food group intake is intake from the dairy food groups (milk, cheese, and yogurt). Adjusted intake is the total daily intake after nutrients from dairy from non-dairy foods (e.g., mixed dishes) have been included, and reflect the disaggregation. Delta intake is the amount of nutrients from dairy in non-dairy foods that was added to or removed from the specific food group intake to calculate the adjusted intake. The consumer number (n) for delta was the number of subjects that consumed dairy from mixed dishes.

Statistical Analyses
Data were analyzed using SAS 9.2 and SUDAAN release 11.0 (Research Triangle Institute, Research Triangle Park, NC, USA) with survey parameters including strata, primary sampling units, and dietary sample weights [26]. Means and standard errors (SE) of energy and nutrient intakes from the total diet and from each food group were determined using PROC DESCRIPT of SUDAAN. Percentages of total energy and nutrient intakes from each food group were calculated from the average consumption of each food. Mean intakes were tabulated by ranked order to 1% of consumption.

Contribution of Foods to Percent Fiber Intake
Total mean daily dietary fiber intake was 11.8 ± 0.2 g; 14.6 ± 0.3 g; and 14.7 ± 0.3 g, for children 2-5, 6-11, and 12-18 years, respectively. Table 2 shows the food sources contributing to at least 1% of daily fiber intake. In all three age groups there were 22 different food groups that contributed at least 1% of fiber intake. For the specific food group intakes, fruit was the top contributor to fiber intake with a mean of 2 g; 17.2%; 1.9 g; 12.8%, and 1.5 g; 10%, for each age group, respectively. For children 2-5 years, SFG intake was followed by bread, rolls, tortillas (1.2 g; 10%) and ready-to-eat cereal (RTEC) (0.9 g; 7.4%). For children 6-11 years, bread, rolls, tortillas (1.5 g; 10%) and mixed dishes-pizza (1.1%; 7.4%) followed; finally, in children 12-18 years bread, rolls, tortillas (1.5 g; 10%), was followed by mixed dishes-Mexican (1.1 g; 7.7%). There were no differences in rank order after adjustment of any of the foods in all three age groups and delta intake was zero.

Discussion
This study showed that top food sources contributing to intake of energy, fiber, calcium, vitamin D, potassium, SFA, added sugars, and sodium varied by age group. In addition, food groups providing some of the major sources of nutrients of public health concern also contributed nutrients to limit in the diet. Mixed dishes, especially pizza and Mexican dishes, contributed to the intake of short fall nutrients in the diets of children.
Nutrients of public health concern [2] have been identified as the shortfall nutrients that pose a substantial risk to the health of our nation. In adults, fiber intake has been associated with a protective effect against gastrointestinal diseases, obesity, CVD, and type 2 diabetes [35]. Fewer studies have been conducted in children; thus, the full impact of dietary fiber intake by children is not clear [36]. Calcium has long been associated with bone and tooth health, but it has also been associated with reduction in the risk of CVD and hypertension; cancers of the colon, rectum, and prostate; kidney stones; and weight management [37]. Potassium is perhaps best recognized for its effect on lowering/controlling blood pressure [38], but other health effects of low potassium intake include a higher risk of stroke [39], insulin resistance, and diabetes [40]. For vitamin D consumption, this study assessed vitamin D3, a prohormone produced in skin through ultraviolet irradiation of 7-dehydrocholesterol, and vitamin D2, found principally in plants [41]. Vitamin D increases intestinal calcium and phosphate absorption, bone calcium mobilization, and the renal reabsorption of calcium, thereby supporting bone mineralization and preventing nutritional rickets in children and osteomalacia in adults [41]. Vitamin D also has other physiologic function, including modulating cell growth, neuromuscular and immune functions, and reducing inflammation [37].
For the nutrients of public health concern [2], dairy products, particularly, milk, provided the top source of calcium, vitamin D, and potassium for all age groups. Although dairy foods provided the top sources of most of these nutrients for most age groups in the SFG data, when the data were adjusted, mixed dishes that included dairy products contributed substantially to intake of these nutrients. This suggested that these foods were no longer important sources of calcium, vitamin D, and potassium. For example, in children 6-11 years, the SFG data showed that mixed dishes-pizza was the 3rd top source of calcium; however, after adjustment, mixed dishes-pizza fell to the 20th source as the nutrients from cheese on pizza were reassigned to the cheese food group. In children 12-18 years, mixed dishes-pizza went from the 3rd top source of calcium to the 23rd top source after adjustment. Thus, it is important to recognize that, food groups that contribute nutrients to limit to the diet can also contribute significantly to the intake of nutrients of public health concern.
Although other foods including most dairy foods provide calcium, milk is well established as the principal source of calcium intake by children [3]. The present study, which used disaggregated data, however, clearly showed that dairy foods in mixed dishes and other dishes contributed many of the nutrients of public health concern found in milk. When examining food sources of nutrients in children these other foods should be considered. Milk and other dairy foods are commonly considered to be an important source of dietary potassium and the recommendation changed to 3 CE/day for most age groups in 2005, in part to increase potassium intake [42].
These data clearly demonstrated the importance of fortification of foods with vitamin D. Using adjusted data, fortified foods contributed 75, 72, and 68% of vitamin D intake by the three age groups; with milk/flavored milk contributing the highest amount of dietary vitamin D. These data contrast sharply with a recent study of children in Ireland [43], where milk/yogurt contributed only 13% of dietary vitamin D since most milk in that country is not fortified [43,44]. Since vitamin D increases calcium absorption, the combination of vitamin D and calcium is especially important for bone health. Fortification of milk has been recently reviewed [45]. The 2015-2020 DGAC [2] reconfirmed that vitamin D is a nutrient of public health concern. Data from WWEIA 2013-2014, showed that the intake of vitamin D by children 2-19 years was only 244 IU [13], which is less than half of the 600 IU dietary reference intake recommendation for this age group [37]. Fortification of foods, especially milk and RTEC, is an important way to increase dietary intake of vitamin D. These foods, as well as other foods high in vitamin D, including egg yolks and salmon, should be encouraged. The importance of milk and other dairy foods to potassium intake was clear, as it provided the top source of the nutrient in all three age groups. Potassium intake is very low, with average intakes slightly over half the requirements of most children [13,24]. Thus, high potassium foods including milk and other dairy foods should continue to be encouraged, along with other high potassium foods, notably fruit and vegetables.
Despite the contribution of dairy foods to the intake of shortfall nutrients, including in mixed dishes, there is concern that they contribute high amounts of SFA and sodium to the diet. Using adjusted data, for children 2-5 years, milk, flavored milk, and cheese contributed 39% of SFA (yogurt and dairy drinks/substitutes contributed minimally to SFA in the diet); for children 6-11 milk, flavored milk, and cheese contributed 31% of SFA to the diet; and for children 12-18 years 32% of SFA came from these dairy foods. The DGAC [2] recommends that no more than 10% of energy come from SFA. In this study, the amount of energy from SFA consumed by children varied by age; for children 2-5, 6-11, and 12-18 years the percent energy from SFA was 11.8, 12.0, and 11.4, respectively.
The rationale for the current recommendation is that by reducing SFA, low-density lipoprotein cholesterol levels are reduced and, in turn, the risk for CVD is lowered. Cardiovascular disease, which is the principal cause of death in the world, has its roots in childhood [46]. Recently, however, the relationship between SFA and CVD have been questioned [47][48][49][50], in part due to the nutrients that would replace SFA in the diet [51] and in part because not all food sources of SFA are associated with an unfavorable risk of CVD [52]. A number of studies, including several meta-analyses have shown that consumption of dairy products is associated with a neutral or inverse risk of CVD [53][54][55].
One of the easiest ways to reduce the amount of SFA in the diets of children two years and older is to encourage the consumption of low fat milk or flavored milk. When the category description of the milk sub-group was examined, low fat milk contributed only 0.25 g of SFA to the overall intake (data not shown). However, other sources of SFA in the diet also need to be addressed, notably sweet bakery products, such as cookies, brownies, and doughnuts. These foods are also among the principal sources of added sugars to the diet. Reducing the SFA and added sugars content of these foods is more difficult than for milk, since some of the structural integrity and sensory properties of these products are linked to solid fat [56,57]. Thus, consumer education may be the best way to reduce the intake of SFA in the diet.
Cheese was the principal contributor of sodium to the diet in the two older groups of children. Analysis of the contributions of disaggregated food mixtures showed this more clearly than examining the foods "as consumed". In addition to cheese consumed directly, cheese was an important ingredient in mixed dishes widely consumed by children, including pizza and Mexican dishes. Thus, reducing the sodium in the diet by reducing the amount of cheese consumed may prove difficult. Reformulating cheeses as reduced-sodium products may also be challenging. Not only does salt help prevent microbial growth in cheese [58], low sodium cheeses may not be well received by consumers [59]. A gradual reduction in the amount of salt used in cheese manufacture may help introduce consumers to a lower sodium product [59] or replacement of part of the sodium chloride with potassium chloride [60]. Reduction of cheese in the diet may be an option; however, this would limit the intake of other shortfall nutrients found in cheese.
This study had a number of strengths. The first is was that it used a large, nationally representative sample. The study also demonstrated the differences in food sources of nutrients in three age groups of children. The third is was that disaggregated energy and nutrients from milk, cheese, yogurt, and non-dairy food groups were also considered which gives further insight into the relative contribution of milk, cheese, and yogurt to both nutrients to encourage and to limit. In addition, this approach can help individuals make more informed food choices [61].
The study also had a number of limitations. A limitation is was the use of 24-h dietary recalls to assess intake in NHANES. Participants or proxies relied on memory to self-report dietary intakes; therefore, data were subject to non-sampling errors, including under or over-reporting of energy and foods. The proxies reporting for or assisting children 2-11 years may not know what their children consumed outside of the home [62], which could also result in reporting errors [63]. Concerns about the validity of self-reported dietary intakes in NHANES has led to an ongoing debate about the validity of these data. Some believe strongly that the data are virtually useless [64][65][66] given issues with misreporting, whereas others, including the prestigious National Cancer Center, [2,67,68] use the data recognizing any potential limitations and allow conclusions to be drawn accordingly. According to Ahluwalia [68], the Nutrition Monitoring Advisor for the Division of Health and Nutrition Examination Surveys, NCHS, Centers for Disease Control and Prevention, and coworkers "NHANES collects dietary data in the context of its broad, multipurpose goals". Their recent review discusses further strengths and limitations of these data. Finally, it should be remembered that cross-sectional studies are used to generate hypotheses, not to test them.
The question may arise as to why "added sugars", as defined by the Dietary Guidelines for Americans [42] were used in this study rather than the "free sugars" designation used by the World Health Organization [69]. These two terms differ significantly since "free sugars" include those sugars naturally occurring in " . . . fruit juice". The authors do not believe that 100% fruit juice, which by definition, as no sugar added should not be considered in the category of free sugars. When evidence-based studies were examined, 100% fruit juice has consistently been shown not to contribute to overweight or obesity in children [70][71][72] or adults [73], instead contributing to nutrient intake and nutrient adequacy, and higher diet quality [71,[73][74][75]. Furthermore, since we worked with an American population, it was felt that using the definitions provided by the nutrition policy statement of the US government was more appropriate.
Due to the technical difficulties involved, dairy was disaggregated from mixed dishes only; further insights might be obtained if it had been possible to disaggregate other food groups. Lastly, for this study, the assumption was made that the milk, cheese, and yogurt components of a mixed food follow the nutrient profiles of milk, NFS; cheese, NFS; and yogurt, NFS, but this approach may not provide the best approximation for some foods. For example, some types of cheese in a mix dishes may deviate from having a nutrient content similar to 'cheese, NFS' for one or more nutrients.

Conclusions
This study showed that for children in all three age groups studied, mixed dishes containing dairy foods contributed to calcium, vitamin D, and potassium intake-three of the nutrients of public concern. A caveat of dairy food consumption is that full fat dairy can contribute saturated fatty acids to the diet and cheese, a major component of many of the mixed dishes, such as pizza and Mexican foods, contributes not only saturated fatty acids, but sodium to the diet. The study also showed that fortifying foods with vitamin D was important since few foods contain naturally occurring vitamin D. The study also showed that children, especially those 6-11 and 12-18 years consumed a large proportion of total energy from energy-dense low-nutrient food groups, such as sugar sweetened beverages and sweet bakery products. Those foods contributed little to the nutrients of public health concern, but did contribute to the nutrients to limit, notably added saturated fatty acids and added sugars. Awareness of food and beverage sources of nutrients can help health professionals design and promote effective age-appropriate strategies to increase the nutrient density of the diet. In addition, this awareness can help the food industry to design and market foods frequently consumed by children that are acceptable and lower in energy and nutrients to limit. Partial support was also received from the National Dairy Council. The sponsors had no input into the design, analyses, or interpretation of the results; and did not read the final manuscript prior to submission.

Conflicts of Interest:
Aside from the information above on funding support, the authors declare no other conflicts of interest.