Food Processing, According to the Nova Classification System, and Dietary Intake of US Infants and Toddlers

Abstract

Background: High intake of ultra-processed foods (UPFs) is associated with increased risk of chronic disease; thus, it is important to understand how UPFs influence diet quality early in life.

Objectives: We describe complementary foods and beverages (CFBs) according to the Nova Classification System of Food Processing for infants and toddlers in the United States and estimate how Nova groups and subgroups contribute to energy and select nutrients and food groups.

Methods: We used day 1 24-h recall from infants and toddlers aged 6-23 mo from the cross-sectional, nationally representative 2013-18 National Health and Nutrition Examination Survey (n = 1140). We estimated contributions of Nova groups and subgroups to energy and select nutrients and food groups consumed as CFBs (excluding human milk and formula) using the population ratio with weighted survey commands in SAS.

Results: For infants and toddlers in the United States, 42 ± 0.9% (mean ± standard error of the mean) of energy intake from CFBs came from unprocessed/minimally processed foods (U/MPFs) and 45 ± 0.8% from UPFs. U/MPFs contributed most to nutrient intakes (except iron, zinc, and sodium); ≥20% of all selected nutrients was from UPFs. UPFs contributed most to iron (75 ± 1.0%) and zinc (48 ± 1.3%); breakfast cereals were the top source. Most fruit, vegetables, and dairy were from U/MPFs. More than 80% of total grains, whole grains, refined grains, and added sugars were UPFs.

Conclusions: U/MPFs support healthy dietary intake of infants and toddlers in the United States, whereas UPFs contribute meaningfully to nutrients and food groups to be encouraged (iron, zinc, and whole grains), as well as some that should be limited (added sugars and sodium). More research is needed to better understand the utility and sensitivities of using Nova for providing dietary guidance for infants and toddlers in the United States.

Keywords: birth to 24-mo; dietary guidance; dietary patterns; food processing; national nutrition surveillance.

FIGURE 1

Participant flowchart for final analytical sample.

FIGURE 2

Percent of energy from complementary foods and beverages consumed by infants and toddlers aged 6–23 mo old in the United States classified according to the Nova Classification System of Food Processing, stratified by demographic characteristics. Data are shown as means ± SEM (n = 1140) using the population ratio method multiplied by 100; e.g., percentage of energy from UPFs = (population sum of kcal from UPFs)/(population sum of total kcal from all complementary foods and beverages, excluding human milk and formula). Data source: United States CDC/NCHS, NHANES 2013–2018. Corresponding summary data and statistical comparisons are shown in Supplementary Table 2. All estimates are deemed reliable according to the NCHS data standards, i.e., a relative SE (calculated by dividing the SE of the estimate by the estimate itself and multiplying by 100) <30% [51], unless otherwise noted in Supplementary Table 2. We also required reliable estimates to have ≥30 reports of each food or beverage (on the food code level) per stratum estimate. Participants were excluded from stratified analyses if the stratified covariate of interest had missing data. Formula-fed is defined as consumed formula, and no human milk, and human milk-fed is defined as consumed human milk and no formula. Those who consumed both or neither are not presented because of the inadequate sample sizes. HS, High school education; NH, non-Hispanic; PIR, Poverty Income Ratio; UPF, ultra-processed food.

FIGURE 3

Percent of energy from complementary foods and beverages consumed by infants and toddlers aged 6 to 23 mo old in the United States classified according to subgroups of the Nova Classification System of Food Processing. Data are shown as means ± SEM (n = 1140) using the population ratio method multiplied by 100; e.g., percentage of energy from ‘Breads’ = (population sum of kcal from ‘Breads’)/(population sum of total kcal from all complementary foods and beverages, excluding human milk and formula). Data source: United States CDC/NCHS, NHANES 2013–2018. Corresponding summary data are available in Supplementary Table 3, as well as results stratified by age group. All estimates are deemed reliable according to the NCHS data standards, i.e., a relative SE (calculated by dividing the SE of the estimate by the estimate itself and multiplying by 100) <30% [51], unless otherwise noted in Supplementary Table 3. We also required reliable estimates to have ≥30 reports of each food or beverage (on the food code level) per stratum estimate. Participants were excluded from stratified analyses if the stratified covariate of interest had missing data.

FIGURE 4

Nutrient intake from complementary foods and beverages consumed by infants and toddlers aged 6–23 mo old in the United States according to the Nova Classification System of Food Processing. Data are shown as means ± SEM (n = 1140) using the population ratio method multiplied by 100; e.g., percentage of iron from UPFs = (population sum of mg of iron from UPFs)/(population sum of total mg iron from all complementary foods and beverages, excluding human milk and formula). Data source: United States CDC/NCHS, NHANES 2013–2018. Corresponding summary data are available in Supplementary Table 4. All estimates are deemed reliable according to the NCHS data standards, i.e., a relative SE (calculated by dividing the SE of the estimate by the estimate itself and multiplying by 100) <30% [51], unless otherwise noted in Supplementary Table 4. We also required reliable estimates to have ≥30 reports of each food or beverage (on the food code level) per estimate. UPF, ultra-processed food, Vit, vitamin.

FIGURE 5

Food group intake from complementary foods and beverages consumed by infants and toddlers aged 6–23 mo old in the United States according to the Nova Classification System of Food Processing. Data are shown as means ± SEM (n = 1140) using the population ratio method multiplied by 100; e.g., percentage of added sugars from UPFs = (population sum of g of added sugars from UPFs)/(population sum of total g of added sugars from all complementary foods and beverages, excluding human milk and formula). Data source: United States CDC/NCHS, NHANES 2013–2018. Corresponding summary data are available in Supplementary Table 4. All estimates are deemed reliable according to the NCHS data standards, i.e., a relative SE (calculated by dividing the SE of the estimate by the estimate itself and multiplying by 100) <30% [51], unless otherwise noted. We also required reliable estimates to have ≥30 reports of each food or beverage (on the food code level) per estimate. UPF, ultra-processed food.