Dietary Micronutrient Intake and the Prevalence of Metabolic Conditions among Children from the United States–Affiliated Pacific Region in the Children’s Healthy Living Program

Abstract

Background

Nutritional intake during childhood can shape health and well-being throughout life. Although excess macronutrient intake is considered the main driver of obesity development, micronutrients, i.e., minerals and vitamins, can potentiate or ameliorate pathological processes of adiposity. Hence, the micronutrient intake relationship to childhood obesity can guide precision approaches to nutritional needs, considering the dietary habits of a population. Childhood obesity is a health disparity throughout the United States–Affiliated Pacific (USAP) region.

Objectives

The study examined the association between micronutrient intake with body mass index (BMI in kg/m²) and the presence of insulin resistance proxy, acanthosis nigricans (AN), in 3529 children aged 2–8 y from the USAP region in the Children’s Healthy Living study.

Methods

The association of micronutrient intakes with BMI and the presence of AN was stratified by World Bank income groups. Main food sources for micronutrients were also identified from 2 d of food records. Obesity and AN were measured by standardized staff.

Results

Most USAP children did not meet daily intake recommendations for micronutrients, with low intake of calcium, potassium, vitamin D, vitamin E, excess sodium, vitamin A, folate, and niacin. Obesity was directly associated with thiamin intake and inversely associated with selenium intake. AN was inversely associated with calcium, copper, iron, phosphorus, potassium, riboflavin, vitamin B6, vitamin D, and vitamin E intakes and directly associated with selenium and pantothenic acid intake. Micronutrient intake imbalances were most associated with insulin resistance and obesity in lower-middle and high-income groups in the USAP region, respectively.

Conclusions

The profile of micronutrient intake in USAP children and its association with obesity and insulin resistance can be used to provide precision nutrition policy guidance according to the World Bank income group to improve micronutrient intake and curb childhood obesity.

Introduction

Nutritional intake during childhood can shape health and well-being throughout life while also preventing adverse health outcomes such as obesity [1,2]. Excess intake of energy-yielding macronutrients such as carbohydrates and lipids can increase body weight and adiposity [3], elevating the risk of obesity development. Paradoxically, obesity can be characterized by a “hidden hunger,” in which a high-energy intake diet does not provide nutritional adequacy of other dietary nutrients, particularly micronutrients, i.e., vitamins and minerals [4,5]. The dietary intake of macronutrients and micronutrients can modulate hormonal and cellular processes related to adiposity and energy metabolism, influencing energy use in ways that may prevent, mitigate, accelerate, or delay the development of obesity [[6], [7], [8], [9], [10], [11], [12]], particularly during childhood growth [[13], [14], [15]]. An imbalance of micronutrient intake, whether deficient or excessive, can adversely affect human health. Children are particularly vulnerable to disturbances in micronutrient intake, as their bodies are not yet fully developed, and both deficiencies and toxic concentrations of micronutrients can curb growth and increase a child’s susceptibility to obesity [[16], [17], [18], [19], [20], [21]]. For example, selenium deficiency has been linked to myopathies and cardiomyopathies in adults and children [22], whereas excess selenium increases the risk for the development of type 2 diabetes and certain cancers in adults [[23], [24], [25], [26]]. In the United States, the recommended dietary allowance of selenium for children up to age 8 ranges from 15 to 30 μg/d, compared to a recommended dietary allowance of 55 μg/d for adults. Thus, children are at greater risk of selenium toxicity if exposed to the same quantity and food items as adults. Given this example, it is imperative to learn if the micronutrient intake of the child population is adequate and age-appropriate, providing full benefits to health and laying the foundation for appropriate dietary decision-making both individually and for the population. This need is especially pronounced for underrepresented and vulnerable groups, such as in the United States–Affiliated Pacific (USAP) region.

Natural foods vary widely in micronutrient content, usually reflecting characteristics of source plants and animals, including cellular mechanisms to acquire, metabolize and retain each micronutrient [27], age and health status [28,29], the surrounding soil composition [30], and processing conditions [31,32]. The obesogenic nature of nutrient intake in children depends on many factors, including individual variability in metabolizing micronutrients, diversity in biological mechanisms to store and process micronutrients in different food items, and diversity of food sources, the latter commonly reflecting a family’s income and education status [33] which in turn influences consumption. Moreover, a healthy dietary intake that can prevent obesity is an intricate combination of factors permeated by cultural [34,35] and geographic factors [36,37], with local availability dictating dietary intake. Evaluating micronutrient intake adequacy remains challenging due to population-specific factors, necessitating localized measurements and targeted efforts to achieve positive nutritional outcomes [38].

Micronutrient intake of the people living in the USAP region, which encompasses the jurisdictions of Alaska, Hawaiʻi, Guam, American Samoa, Republic of Palau, Commonwealth of the Northern Mariana Islands (CNMIs), Republic of Marshall Islands (RMIs), and the Federated States of Micronesia (FSM; composed of Chuuk, Pohnpei, Kosrae, and Yap states), has been investigated in adults, commonly with a focus either on a certain location or on specific vitamins or minerals [39]. In children, previous dietary intake assessments in Kiribati and New Zealand, both nonaffiliated with the United States, detected micronutrient deficiencies [40,41], whereas a pooled cohort from the East Asia and Pacific group that included the FSM also reported micronutrient deficiencies without specifying micronutrient concentrations for the FSM population [42]. Data on micronutrient intake, in addition to socio-economic and cultural factors that influence dietary intake for the USAP population, are needed to develop precision nutrition guidance that will impact these underserved communities [43]. Moreover, the association of micronutrient intake with nutritional status markers, such as obesity, has not been evaluated in USAP children, accounting for their jurisdiction-specific economic disparities.

For most micronutrients, it is currently unknown if intake is adequate, both in children and adults, in the USAP region, overall, or in relation to the regional socio-economic status. In addition, it is unclear whether local foods are the primary sources of micronutrients present in the Pacific. For example, seafood is heavily consumed regionally due to its easy access in most islands and could be a major source of dietary micronutrients in Pacific communities, including the jurisdictions evaluated by the Children’s Healthy Living (CHL) program [44].

Therefore, a comprehensive systematic analysis of food sources consumed by the specific communities of the USAP region, their micronutrient content, and their association with obesity prevalence or the presence of acanthosis nigricans (AN) as a proxy for insulin resistance was performed. The hypothesis is that inadequate (i.e., deficient and/or excessive) micronutrient intake in the diet consumed by the USAP region children is associated with the increased prevalence of overweight and obesity (OWOB) and the presence of AN, both conditions being potentially amplified in the younger members of the population. The study utilized dietary record data obtained from the CHL study that recorded the reported intake of foods and beverages of children from communities in the USAP region [44]. The intake was evaluated against food composition data for intake of vitamins (choline, folate, niacin, pantothenic acid, riboflavin, thiamin, vitamin A, vitamin B6, vitamin B12, vitamin C, vitamin D, and vitamin E) and minerals (calcium, copper, iron, magnesium, manganese, phosphorus, potassium, selenium, sodium, and zinc). Micronutrient intake values were examined in relation to OWOB and classified based on BMI measurements according to age and sex. They were compared to the reference and the presence or absence of AN and tested by the World Bank income level of each USAP jurisdiction.

Methods

The CHL program included a multijurisdictional prevalence study and intervention trial (clinicaltrial.gov identifier: NCT01881373) for children 2–8 y old in the USAP region. Details about the study are given elsewhere [44]. Cross-sectional data, including demographic information, 2 d of food and activity records, and anthropometric measurements were collected by trained staff from 24 selected communities across 11 jurisdictions (Hawai‘i, Alaska, CNMI, Guam, American Samoa, Palau, RMI, and the 4 FSM: Pohnpei, Yap, Kosrae, and Chuuk) in a community cluster design [44]. A total of 4178 children participated in the CHL prevalence study at baseline. By design, a subsample of 3529 children provided 2 nonconsecutive days of food records, the sample for this study.

Demographics

Demographic information reported by caregivers included household income and the child’s age, sex, and race or ethnicity according to the Office of Management and Budget categories [45]. Additional race and ethnicity subcategories were provided under Asian, Native Hawaiian and Pacific Islanders, and American Indian/Alaska Native, including a write-in opportunity. An indigenous variable was developed from the Pacific Islander and Alaskan Native ethnicities, indicating the ethnicity matching the jurisdiction in which the data was collected (Palau: Palauan, Yap: Yapese, Guam: CHamoru, CNMI: CHamoru and Carolinian, Chuuk: Chuukese, Pohnpei: Pohnpeian, Kosrae: Kosraean, RMI: Marshallese, American Samoa: Samoan, Hawai‘i: Native Hawaiian, Alaska: Alaskan Native or Pacific Northwest American Indian). World Bank Income Groups (WBIG) criteria from the year the data was collected [46] was used, and these criteria establish that gross national income per capita (GNIPC) stratifies country economies into 4 categories: low-income economies were those with a GNIPC of $1045 or less; lower-middle-income (LMI) economies were those with a GNIPC between $1046 and $4125; upper-middle-income (UMI) economies were those with a GNIPC between $4126 and $12,735; high-income (HI) economies were those with a GNIPC of $12,736 or more. The CHL jurisdictions were classified into 3 economic groups: LMI jurisdictions (Chuuk, Pohnpei, Yap, and Kosrae, all members of the FSM), UMI jurisdictions (Palau, American Samoa, and Marshall Islands), and HI jurisdictions (Alaska, Hawaiʻi, Guam, and Northern Marianas).

Health outcomes

Anthropometry, including weight (in kilograms) and height (in centimeters), was measured for CHL participants by a trained and standardized staff member, adhering to a standard protocol [47]. Three measurements were taken in total and means were calculated for analysis. BMI percentiles were determined by the Centers for Disease Control and Prevention, where underweight was <5th percentile, healthy weight was 5th to the 84th percentile, overweight was 85th to the 94th percentile, and obesity was >95th percentile [48], with BMI adjustments made for age and sex. AN was determined by assessing the back of the neck using the Burke method with a scale of 0–4 and analyzed as absent (0) or present (1–4) [49].

Micronutrient intake

The variability in micronutrient intake and food sources of vitamins and minerals were investigated from dietary records collected from a subsample of 3529 children in the CHL study, completed by a caregiver as previously described [50]. Micronutrient intakes were assessed using 2 systematically assigned days of food records, records which were completed by caregivers of the children in the CHL program. This method was selected to ensure a comprehensive representation of dietary patterns across all days of the week to reduce variability and bias in reporting. Measuring tools were provided to caregivers, and wrappers, labels, and packages were collected. Standard techniques were used to improve the accuracy of information recorded in the food records [51]. Dietary records were entered by trained staff into a web-based Pacific tracker (PacTrac) version 3.1 to evaluate the dietary record data from the Pacific region [52]. PacTrac includes a food composition table, developed and validated by the University of Hawaiʻi Cancer Center [53,54], that accounts for a wide range of food commonly consumed in the Pacific region, including indigenous foods and those typical of American, Asian, and Alaskan diets. The tool provides an accurate estimation of nutrient intake based on the United States Dietary Guidelines and the Healthy Eating Index. To minimize potential bias, the dietary components were averaged across days, with adjustments made to account for weekday and weekend days, and corrected for within-person variance [55]. Estimated Average Requirements (EAR) values were used to evaluate micronutrient intake, or adequate intake (AI) when no EAR was available, to determine the status of meeting recommendations based on age and sex [56] (Table 1 and Supplementary Table 1). The status of exceeding the tolerable upper limit (UL) was also identified, based on the chronic disease risk reduction level if UL was unavailable, which was the case for sodium. The top 5 most important food contributors to the micronutrient intake among consumed food items were identified based on a percentage of the micronutrients derived from that specific food item in the population.

TABLE 1.

Description of the study sample by Pacific jurisdiction and World Bank income groups, the children’s healthy living program (2013–2014).

Characteristic	Lower-middle income				Upper-middle income			High income				Total
	Chuuk	Kosrae	Pohnpei	Yap	American Samoa	Palau	RMI	Alaska	CNMI	Guam	Hawai‘i
Total, n	123	129	187	190	588	166	191	313	546	666	430	3529
Sex, n (%)
Male	70 (56.9)	72 (55.8)	91 (48.7)	88 (46.3)	306 (52)	89 (53.6)	82 (42.9)	175 (55.9)	283 (51.8)	339 (50.9)	203 (47.2)	1798 (51)
Female	53 (43.1)	57 (44.2)	96 (51.3)	102 (53.7)	282 (48)	77 (46.4)	109 (57.1)	138 (44.1)	263 (48.2)	327 (49.1)	227 (52.8)	1731 (49.1)
Age group, n (%)
2–3 y	1 (0.8)	0	22 (11.8)	24 (12.6)	148 (25.2)	29 (17.5)	3 (1.6)	90 (28.8)	106 (19.4)	129 (19.4)	147 (34.2)	699 (19.8)
4–8 y	122 (99.2)	129 (100)	165 (88.2)	166 (87.4)	440 (74.8)	137 (82.5)	188 (98.4)	223 (71.3)	440 (80.6)	537 (80.6)	283 (65.8)	2830 (80.2)
Indigenous1, n (%)
Yes	121 (98.4)	126 (97.7)	183 (97.9)	186 (97.9)	579 (98.5)	157 (94.6)	191 (100)	70 (22.4)	320 (58.6)	444 (66.7)	320 (74.4)	2697 (76.4)
No	2 (1.6)	3 (2.3)	4 (2.1)	4 (2.1)	9 (1.5)	9 (5.4)	0	243 (77.6)	226 (41.4)	222 (33.3)	110 (25.6)	832 (23.6)
Income per year, n (%)
<$35,000	37 (30.1)	99 (76.7)	121 (64.7)	108 (56.8)	450 (76.5)	149 (89.8)	98 (51.3)	153 (48.9)	389 (71.3)	334 (50.2)	215 (50)	2153 (61)
≥$35,000	2 (1.6)	0	8 (4.3)	0	27 (4.6)	8 (4.8)	7 (3.7)	139 (44.4)	36 (6.6)	134 (20.1)	172 (40)	533 (15.1)
Missing	84 (68.3)	30 (23.3)	58 (31)	82 (43.2)	111 (18.9)	9 (5.4)	86 (45)	21 (6.7)	121 (22.2)	198 (29.7)	43 (10)	843 (23.9)
BMI category, n (%)
Underweight	5 (4.1)	5 (3.9)	0	12 (6.3)	2 (0.3)	4 (2.4)	14 (7.3)	2 (0.6)	24 (4.4)	19 (2.9)	12 (2.8)	99 (2.8)
Healthy weight	107 (87)	110 (85.3)	139 (74.3)	152 (80)	330 (56.1)	121 (72.9)	171 (89.5)	204 (65.2)	380 (69.6)	463 (69.5)	280 (65.1)	2457 (69.6)
Overweight	9 (7.3)	8 (6.2)	30 (16)	12 (6.3)	105 (17.9)	11 (6.6)	5 (2.6)	61 (19.5)	54 (9.9)	84 (12.6)	71 (16.5)	450 (12.8)
Obese	1 (0.8)	6 (4.7)	7 (3.7)	8 (4.2)	149 (25.3)	26 (15.7)	0	38 (12.1)	87 (15.9)	87 (13.1)	65 (15.1)	474 (13.4)
Missing	1 (0.8)	0	11 (5.9)	6 (3.2)	2 (0.3)	4 (2.4)	1 (0.5)	8 (2.6)	1 (0.2)	13 (2)	2 (0.5)	49 (1.4)
Acanthosis nigricans present, n (%)
Yes	4 (3.3)	4 (3.2)	22 (11.8)	4 (2.1)	47 (8)	11 (6.6)	22 (11.5)	0	51 (9.3)	21 (3.2)	6 (1.4)	192 (5.5)
No	118 (95.9)	125 (96.9)	160 (85.6)	180 (94.7)	540 (91.8)	152 (91.6)	168 (88)	310 (99)	495 (90.7)	640 (96.1)	420 (97.7)	3308 (93.7)
Missing	1 (0.8)	0	5 (2.7)	6 (3.2)	1 (0.2)	3 (1.8)	1 (0.5)	3 (1)	0	5 (0.8)	4 (0.9)	29 (0.8)

Abbreviations: BMI, body mass index; CNMI, Commonwealth of the Northern Mariana Islands; RMI, Republic of Marshall Islands.

¹Indigenous: culturally distinctive ethnicities that originated from specified jurisdiction.

Statistical analysis

Prevalence and means, with SE, were estimated using survey sampling techniques that weighted the sample to the young child population size in each community based on census data and accounted for the clustering of participants in communities within jurisdictions. The weighting resulted in representative estimates for the jurisdictions and the region. In Hawai‘i, the communities largely represented rural areas, and in Alaska, the communities largely represented urban areas. Hierarchical marginal logistic models of prevalence for meeting recommendations, adjusting for age and sex, were fit to compare subgroups and evaluated for statistical significance by the Wald test. Hierarchical marginal linear models of micronutrients (adjusting for age, sex, and dietary energy) were fit to compare means between subgroups. Distributions of residuals were checked to ensure that model assumptions were met; no dietary variable required transformation. Dietary variables were analyzed and compared among WBIGs, between OWOB and healthy-weight children, and between children with and without AN to assess their status. Underweight children (n = 99) were excluded from the analysis of OWOB compared to healthy-weight children but still included in the calculation of the average intake of micronutrients. A P value <0.05 was considered statistically significant. Statistical analysis was done using SAS 9.4 (SAS Institute Inc).

Ethical standard disclosure

This study was conducted and analyzed according to the guidelines of the Declaration of Helsinki. Approval for human studies was obtained from the institutional review board Committee on Human Studies at the University of Hawaiʻi at Mānoa (CHS#18915) and at the University of Guam for data collection and ongoing data analysis (this study). All other participating jurisdictions ceded approval to the University of Hawaiʻi at Mānoa. The written/verbal consent of parents and the assent of children were obtained prior to participation in the study. Participants received $20 for participation in all locations, except in Guam and Alaska, where compensation was $40 and $50, respectively.

Results

The study included a total of 3529 children aged 2–8 y with dietary data, with an even distribution of boys and girls (Table 1). A majority (76.4%) of the sample were indigenous to their respective jurisdictions, except in Alaska. The prevalence of children who were classified as overweight and obese, based on the Centers for Disease Control and Prevention BMI categories, was 12.8% and 13.4%, respectively. A total of 5.5% of children had a presence of AN.

Vitamin intake

Table 2 presents the intake percentage stratified by WBIG of children meeting the daily recommended intake for vitamins and minerals, with a breakdown of the intake of each micronutrient per jurisdiction found in Supplementary Table 2. Overall, the majority of children met vitamin recommendations except for vitamin D and vitamin E. Across WBIGs, a low percentage (10.7–34.1%) of children met recommendations for vitamin D and vitamin E, being low in intake of these vitamins, i.e., with intake below the EAR. Vitamin D intake recommendations for children were met by only 4.3% in LMI, 7.6% in UMI, and 2.2% in HI jurisdictions, whereas vitamin E intake recommendations were met by 10.7% in LMI, 34.1% in UMI, and 17.5% in HI jurisdictions. Additionally, low percentages of children met recommended intakes for choline (LMI: 39%; UMI: 66.2%; HI: 56.6%), folate (LMI: 74.6%; UMI: 42.7%; HI: 40%), niacin (LMI: 25.4%; UMI: 12.2%; HI: 16.2%) and vitamin A (LMI: 46.7%; UMI: 71.3%; HI: 78.3%) despite adequate average intakes across WBIGs. Virtually all children who did not meet recommendations for niacin exceeded UL intakes. A high proportion of children exceeded the UL intake for folate (LMI: 24.6%; UMI: 55.5%; HI: 58.9%) and vitamin A (LMI: 53.3%; UMI: 28.7%; and HI: 21.7%). Recommended intakes for all other vitamins (pantothenic acid, riboflavin, thiamin, vitamin B6, vitamin B12, and vitamin C) were generally met by over 90% of children across WBIGs (Supplementary Figure 1).

TABLE 2.

Meeting micronutrient dietary recommendations among the United States–affiliated Pacific children by World Bank income groups.

Micronutrient	Recommended intake	Meets recommendation	Lower middle (%)	Upper middle (%)	High (%)	Total (%)
Vitamins
Choline	1–3 y: 200–1000 mg/d	Yes	39.0	66.2	56.6	53.9
	4–8 y: 250–1000 mg/d	No	61.1	33.8	43.4	46.1
Folate	1–3 y: 120–300 μg/d	Yes	74.6	42.7	40.0	52.4
	4–8 y: 160–400 μg/d	No - lower limit	0.8	1.8	1.1	1.2
		No - upper limit	24.6	55.5	58.9	46.3
Niacin	1–3 y: 5–10 mg/d	Yes	25.4	12.2	16.2	17.9
	4–8 y: 6–15 mg/d	No - lower limit	0.0	0	0.2	0.1
		No - upper limit	74.6	87.8	83.6	82.0
Pantothenic Acid	1–3 y: 2 mg/d	Yes	92.2	94.2	93.5	93.3
	4–8 y: 3 mg/d	No	7.8	5.8	6.5	6.7
Riboflavin	1–3 y: 0.4 mg/d	Yes	97.6	99.4	99.6	98.9
	4–8 y: 0.5 mg/d	No	2.4	0.6	0.4	1.1
Thiamin	1–3 y: 0.4 mg/d	Yes	97.1	98.5	99.5	98.4
	4–8 y: 0.5 mg/d	No	2.9	1.5	0.5	1.6
Vitamin A1	1–3 y: 210–600 μg RAE/day 5 4–8 y: 275–900 μg RAE/day 5	Yes	46.7	71.3	78.3	65.4
		No - lower limit	0.0	0.0	0.0	0.0
		No - upper limit	53.3	28.7	21.7	34.6
Vitamin B6	1–3 y: 0.4–30 mg/d	Yes	99.0	99.3	99.7	99.3
	4–8 y: 0.5–40 mg/d	No	1.0	0.7	0.3	0.7
Vitamin B12	1–3 y: 0.7 μg/d 4–8 y: 1.0 μg/d	Yes	98.9	99.8	100.0	99.6
		No	1.1	0.2	0.0	0.4
Vitamin C	1–3 y: 13–400 mg/d	Yes	82.8	92.1	96.9	90.6
	4–8 y: 22–650 mg/d	No	17.2	7.9	3.1	9.4
Vitamin D	1–3 y: 10–50 μg/d 4–8 y: 10–50 μg/d	Yes	4.3	7.6	2.2	4.7
		No	95.7	92.4	97.8	95.3
Vitamin E	1–3 y: 5–200 mg/d	Yes	10.7	34.1	17.5	20.8
	4–8 y: 6–300 mg/d	No	89.4	65.9	82.5	79.3
Minerals
Calcium	1–3 y: 500–2500 mg/d	Yes	5.6	37.0	39.5	27.4
	4–8 y: 800–2500 mg/d	No	94.4	63.0	60.5	72.6
Copper	1–3 y: 260–1000 μg/d	Yes	97.5	90.6	94.3	94.1
	4–8 y: 340–3000 μg/d	No - lower limit	0.2	0.5	0.2	0.3
		No - upper limit	2.3	8.9	5.5	5.6
Iron	1–3 y: 3.0–40 mg/d	Yes	99.4	99.9	99.8	99.7
	4–8 y: 4.1–40 mg/d	No	0.6	0.1	0.2	0.3
Magnesium	1–3 y: 65 mg/d	Yes	97.9	97.3	98.0	97.7
	4–8 y: 65 mg/d	No	2.1	2.8	2.0	2.3
Manganese	1–3 y: 1.2–2 mg/d	Yes	94.6	93.1	93.4	93.7
	4–8 y: 1.5–3 mg/d	No	5.4	6.9	6.6	6.3
Phosphorus	1–3 y: 380–3000 mg/d	Yes	99.4	99.3	99.6	99.4
	4–8 y: 405–3000 mg/d	No	0.6	0.7	0.4	0.6
Potassium	1–3 y: 2000 mg/d	Yes	18.0	39.1	28.1	28.4
	4–8 y: 2300 mg/d	No	82.0	60.9	71.9	71.6
Selenium	1–3 y: 17–90 μg/d	Yes	76.6	70.6	86.9	78.0
	4–8 y: 23–150 μg/d	No - lower limit	0.0	0.0	0.2	0.1
		No - upper limit	23.4	29.4	12.90	21.9
Sodium	1–3 y: 800–1200 mg/d CDRR	Yes	15.4	6.0	4.2	8.5
		No - lower limit	3.7	0.7	0.2	1.5
	4–8 y: 1000–1500 mg/d CDRR	No - upper limit	80.9	93.2	95.6	90.0
Zinc	1–3 y: 2.5–7 mg/d	Yes	91.1	70.5	69.6	77.1
	4–8 y: 4–12 mg/d	No - lower limit	7.8	28.4	29.8	22.0
		No - upper limit	1.1	1.1	0.6	0.9

Abbreviations: CDRR, chronic disease risk reduction, RAE, retinol activity equivalent.

¹Vitamin A is μg RAE/day. One μg retinol, 12 μg beta (β)-carotene, and 24 μg alpha-carotene or β-cryptoxanthin [56].

Vitamin intake and health outcomes

Except for thiamin, most vitamin intake had no association with OWOB status in children of the USAP region (Table 3). When considering WBIGs, the average intake of pantothenic acid was statistically higher among children without OWOB compared to children with OWOB in HI jurisdictions, uncovering an inverse association between this micronutrient and OWOB. Although just over half of children in LMI jurisdictions met the recommended vitamin A intake (Supplementary Figure 1), healthy-weight children had statistically lower vitamin A intake compared to those with OWOB (Table 3).

TABLE 3.

Daily average micronutrient intake of children by weight status (healthy compared with overweight and obesity) in the United States–Affiliated Pacific.

Micronutrient	Lower-middle income			Upper-Middle income			High income			Overall
	Healthy	OWOB	P value	Healthy	OWOB	P value	Healthy	OWOB	P value	Healthy	OWOB	P value
	Mean ± SE	Mean ± SE		Mean ± SE	Mean ± SE		Mean ± SE	Mean ± SE		Mean ± SE	Mean ± SE
Vitamins
Choline, mg	239.2 ± 7.09	236.46 ± 6.15	0.56	263.8 ± 7.16	274.3 ± 5.06	0.08	264.43 ± 1.96	262.71 ± 1.46	0.35	255.81 ± 3.31	257.82 ± 2.55	0.44
Folate, μg	296.73 ± 21.33	309.15 ± 22.67	0.20	413.05 ± 13.37	413.33 ± 19.34	0.97	456.7 ± 10.87	457.43 ± 12.76	0.97	388.83 ± 9.61	393.3 ± 10.83	0.57
Niacin, mg	18.5 ± 0.74	18.93 ± 0.76	0.13	19.77 ± 0.16	19.41 ± 0.23	0.11	19.06 ± 0.13	19.11 ± 0.2	0.86	19.11 ± 0.24	19.15 ± 0.27	0.8
Pantothenic acid, mg	4.58 ± 0.07	4.55 ± 0.05	0.57	4.61 ± 0.09	4.66 ± 0.06	0.47	4.64 ± 0.05	4.53 ± 0.05	0.01	4.61 ± 0.05	4.58 ± 0.04	0.38
Riboflavin, mg	1.25 ± 0.1	1.26 ± 0.09	0.75	1.66 ± 0.14	1.85 ± 0.09	0.12	1.85 ± 0.08	1.82 ± 0.08	0.26	1.59 ± 0.09	1.64 ± 0.08	0.18
Thiamin, mg	0.99 ± 0.05	1.01 ± 0.05	0.32	1.2 ± 0.04	1.23 ± 0.04	0.10	1.34 ± 0.03	1.35 ± 0.03	0.40	1.17 ± 0.03	1.2 ± 0.04	0.04
Vitamin A, μg	335.52 ± 20.6	369.9 ± 17.53	0.03	459.08 ± 38.23	499.12 ± 12.12	0.29	521.85 ± 13.47	508.09 ± 11.52	0.18	438.82 ± 12.95	459.04 ± 7.74	0.15
Vitamin B6, mg	1.44 ± 0.05	1.44 ± 0.03	0.78	1.47 ± 0.09	1.6 ± 0.04	0.11	1.59 ± 0.03	1.58 ± 0.04	0.89	5.09 ± 0.17	5.1 ± 0.19	0.94
Vitamin B12, μg	5.08 ± 0.27	4.97 ± 0.38	0.56	5.09 ± 0.26	5.24 ± 0.2	0.49	5.11 ± 0.15	5.09 ± 0.16	0.81	1.5 ± 0.04	1.54 ± 0.03	0.14
Vitamin C, mg	53.07 ± 3.52	52.91 ± 1.96	0.97	60.63 ± 4.29	65.79 ± 2.25	0.31	74.91 ± 1.62	76.57 ± 2.6	0.31	62.87 ± 1.74	65.09 ± 1.2	0.29
Vitamin D, IU	186.58 ± 8.15	177.59 ± 11.52	0.28	224.78 ± 13.1	230.4 ± 7.09	0.66	216.64 ± 3.53	209.98 ± 5.21	0.16	209.33 ± 4.62	205.99 ± 4.71	0.53
Vitamin E, mg	4.31 ± 0.15	4.21 ± 0.25	0.62	4.52 ± 0.24	4.83 ± 0.18	0.18	4.67 ± 0.15	4.68 ± 0.15	0.86	4.5 ± 0.15	4.57 ± 0.16	0.47
Minerals
Calcium, mg	482.87 ± 13.66	459.63 ± 8.5	0.01	597.85 ± 46.3	664.94 ± 22.2	0.20	731.54 ± 19.5	705.5 ± 20.34	0.02	604.09 ± 15.67	610.03 ± 11.75	0.74
Copper, mg	0.9 ± 0.03	0.9 ± 0.03	0.91	0.92 ± 0.04	0.98 ± 0.04	0.19	0.96 ± 0.02	0.97 ± 0.02	0.70	0.93 ± 0.02	0.95 ± 0.02	0.21
Iron, mg	10.28 ± 0.25	10.76 ± 0.36	0.02	12.28 ± 0.32	12.67 ± 0.49	0.27	12.84 ± 0.18	12.73 ± 0.18	0.69	11.8 ± 0.15	12.06 ± 0.21	0.12
Magnesium, mg	218.67 ± 8.85	217.31 ± 7.72	0.60	206.72 ± 7.42	212.33 ± 5.13	0.51	216.05 ± 3.43	214.9 ± 2.58	0.55	213.81 ± 4.13	214.85 ± 3.45	0.73
Manganese, mg	2.79 ± 0.13	2.81 ± 0.09	0.80	2.59 ± 0.06	2.53 ± 0.07	0.31	2.64 ± 0.04	2.61 ± 0.01	0.53	2.67 ± 0.05	2.65 ± 0.04	0.59
Phosphorus, mg	935.07 ± 15.06	912.05 ± 19.02	0.02	981.03 ± 30.15	1005.28 ± 18.94	0.45	1039.88 ± 12.59	1024.8 ± 12.88	0.02	985.33 ± 10.54	980.71 ± 9.41	0.68
Potassium, mg	1895.63 ± 95.85	1822.87 ± 74.69	0.11	1872.68 ± 108.91	2015.61 ± 55.22	0.26	2019.71 ± 30.83	2009.64 ± 30.41	0.59	1929.34 ± 48.22	1949.37 ± 35.46	0.65
Selenium, μg	122.02 ± 2.32	122.7 ± 3.27	0.67	110.64 ± 1.53	103.96 ± 2.12	<.001	100.14 ± 0.64	99.66 ± 0.82	0.57	110.93 ± 0.98	108.77 ± 1.31	0.03
Sodium, mg	2220.3 ± 112.58	2117.06 ± 128.43	<.001	2400.01 ± 44.55	2449.94 ± 27.96	0.26	2565.9 ± 23.27	2623.8 ± 17.86	0.01	2395.4 ± 44.29	2396.93 ± 45.57	0.93
Zinc, mg	7.86 ± 0.19	7.87 ± 0.13	0.95	8.35 ± 0.22	8.76 ± 0.23	0.04	9.6 ± 0.13	9.5 ± 0.13	0.32	8.6 ± 0.13	8.71 ± 0.13	0.25

Abbreviations: OWOB, overweight and obesity; SE, standard error.

Models adjusted for age, sex, dietary energy, and jurisdiction. Bolded italicized numbers indicate statistical significance (P < 0.05) after the Wald test.

Overall, 5.5% of the children in this study presented with AN (Table 1). AN presence was associated with intake of pantothenic acid, riboflavin, vitamin B6, vitamin D, and vitamin E (Table 4). When stratified by WBIGs, the presence of AN was associated with higher concentrations of folate, pantothenic acid, and thiamin and with lower concentrations of vitamin B12, vitamin C, and vitamin E in LMI jurisdictions. In UMI jurisdictions, AN was directly associated with choline intake, and in HI jurisdictions, AN was directly associated with pantothenic acid and inversely associated with riboflavin, vitamin B6, vitamin D, and vitamin E (Table 4).

TABLE 4.

Daily average micronutrient intake of children by occurrence of acanthosis nigricans in the United States–Affiliated Pacific.

Micronutrient	Lower-Middle income			Upper-Middle income			High income			Overall
	No	Yes	P value	No	Yes	P value	No	Yes	P value	No	Yes	P value
	Mean ± SE	Mean ± SE		Mean ± SE	Mean ± SE		Mean ± SE	Mean ± SE		Mean ± SE	Mean ± SE
Vitamins
Choline, mg	238.65 ± 6.84	237.47 ± 8.53	0.90	265.41 ± 6.0	278.72 ± 9.5	0.01	263.6 ± 1.71	266.02 ± 2.76	0.40	255.89 ± 2.85	260.73 ± 4.1	0.17
Folate, μg	294.72 ± 20.87	331.05 ± 21.16	0.02	409.9 ± 14.67	428.88 ± 17.94	0.15	459.65 ± 8.02	374.43 ± 11.41	<.001	388.09 ± 9.19	378.12 ± 10.14	0.18
Niacin, mg	18.49 ± 0.74	19.13 ± 0.94	0.37	19.57 ± 0.16	20.28 ± 0.47	0.24	19.06 ± 0.1	18.18 ± 0.36	0.04	19.04 ± 0.25	19.2 ± 0.38	0.64
Pantothenic acid, mg	4.55 ± 0.07	4.71 ± 0.1	<.001	4.61 ± 0.08	4.69 ± 0.12	0.21	4.6 ± 0.05	4.64 ± 0.08	0.50	4.59 ± 0.05	4.68 ± 0.07	0.01
Riboflavin, mg	1.25 ± 0.15	1.29 ± 0.14	0.43	1.71 ± 0.16	1.68 ± 0.2	0.73	1.85 ± 0.14	1.54 ± 0.14	<.001	1.6 ± 0.14	1.5 ± 0.14	0.02
Thiamin, mg	0.98 ± 0.04	1.1 ± 0.07	0.01	1.2 ± 0.04	1.23 ± 0.04	0.19	1.34 ± 0.03	1.23 ± 0.03	<.001	1.17 ± 0.03	1.19 ± 0.03	0.38
Vitamin A, μg	337.36 ± 20.44	376.26 ± 34.85	0.31	468.21 ± 29.02	480.02 ± 47.82	0.60	522.27 ± 12.61	427.24 ± 18.73	<.001	442.61 ± 9.98	427.84 ± 18.55	0.37
Vitamin B6, mg	1.44 ± 0.09	1.39 ± 0.1	0.45	1.5 ± 0.11	1.53 ± 0.13	0.57	1.59 ± 0.09	1.47 ± 0.09	0.02	5.11 ± 0.2	4.79 ± 0.22	0.01
Vitamin B12, μg	5.1 ± 0.29	4.35 ± 0.23	0.01	5.1 ± 0.25	5.36 ± 0.34	0.07	5.12 ± 0.19	4.66 ± 0.29	0.06	1.51 ± 0.09	1.46 ± 0.09	0.15
Vitamin C, mg	53.48 ± 3.19	45.94 ± 1.49	0.02	61.78 ± 3.17	64.25 ± 6.72	0.63	75.69 ± 1.77	68.63 ± 4.85	0.05	63.65 ± 1.39	59.61 ± 2.65	0.09
Vitamin D, IU	184.92 ± 8.47	182.17 ± 7.49	0.75	225.26 ± 10.11	230.81 ± 17.32	0.60	216.05 ± 3.84	176.34 ± 9.41	<.001	208.74 ± 3.97	196.44 ± 5.52	0.04
Vitamin E, mg	4.32 ± 0.24	3.64 ± 0.26	<.001	4.58 ± 0.28	4.73 ± 0.38	0.46	4.69 ± 0.24	4.42 ± 0.25	<.001	4.53 ± 0.24	4.27 ± 0.25	0.01
Minerals
Calcium, mg	479.11 ± 13.73	451.43 ± 17.26	0.19	616.39 ± 35.38	599.9 ± 67.79	0.70	730.31 ± 19.45	577.53 ± 25.12	<.001	608.6 ± 12.46	542.96 ± 21.12	<.001
Copper, mg	0.9 ± 0.05	0.84 ± 0.05	0.04	0.94 ± 0.06	0.9 ± 0.07	0.36	0.97 ± 0.05	0.94 ± 0.05	0.31	0.94 ± 0.04	0.89 ± 0.05	0.03
Iron, mg	10.28 ± 0.26	10.93 ± 0.27	<.001	12.34 ± 0.37	12.52 ± 0.25	0.52	12.87 ± 0.14	11.02 ± 0.18	<.001	11.83 ± 0.15	11.49 ± 0.12	0.02
Magnesium, mg	218.73 ± 8.56	204.86 ± 4.86	0.02	207.91 ± 5.61	207.75 ± 11.17	0.98	215.75 ± 3.23	209.69 ± 7.09	0.42	214.13 ± 3.68	207.43 ± 4.66	0.11
Manganese, mg	2.78 ± 0.12	2.84 ± 0.11	0.64	2.57 ± 0.05	2.55 ± 0.14	0.88	2.62 ± 0.03	2.73 ± 0.1	0.30	2.66 ± 0.04	2.71 ± 0.07	0.45
Phosphorus, mg	931.85 ± 15.39	893.3 ± 29.06	0.17	984.82 ± 22.53	998.81 ± 45.15	0.58	1037.52 ± 12.67	959.81 ± 17.17	<.001	984.73 ± 8.46	950.64 ± 16.7	0.02
Potassium, mg	1896.54 ± 92.76	1639.52 ± 62.82	0.02	1914.45 ± 83.81	1875.04 ± 145.65	0.70	2022.43 ± 30.51	1908.88 ± 54.19	0.03	1944.47 ± 41.36	1807.81 ± 51.12	0.01
Selenium, μg	121.65 ± 2.37	126.21 ± 4.05	0.18	108.4 ± 1.65	111.62 ± 2.04	0.16	99.55 ± 0.53	104.77 ± 1.4	<.001	109.87 ± 0.97	114.2 ± 1.48	0.01
Sodium, mg	2211.5 ± 115.17	2117.87 ± 70.25	0.11	2411.15 ± 35.2	2427.08 ± 78.85	0.77	2579.37 ± 19.16	2579.41 ± 67.38	1.00	2400.68 ± 43.16	2374.79 ± 42.61	0.43
Zinc, mg	7.82 ± 0.28	8.34 ± 0.26	0.02	8.43 ± 0.3	8.63 ± 0.45	0.43	9.58 ± 0.25	8.93 ± 0.26	<.001	8.61 ± 0.25	8.63 ± 0.27	0.86

Abbreviations: SE, standard error.

Models adjusted for age, sex, dietary energy, and jurisdiction. Bolded italicized numbers indicate statistical significance (P < 0.05) after Wald test.

Top food sources of vitamins

Table 5 highlights the top food sources of vitamins across WBIGs, revealing some differences in the foods contributing to vitamin intake. Although there is variability in meeting intake recommendations among WBIGs, canned tuna (20.9%) was the leading source of vitamin D, which is essential for calcium regulation and bone health [57]. In contrast, reduced-fat milk was the primary source of vitamin D in UMI and HI jurisdictions, contributing 17.5% and 26.2%, respectively. Similarly, meeting recommendations for niacin was somewhat higher in LMI at 25% compared to UMI at 12% and HI at 16%, niacin was commonly sourced by canned tuna (10.2% and 7.4%) in LMI and UMI, and enriched white rice (3.3%) in HI jurisdictions. For choline, saimin noodles (5.4%) in LMI, hard-cooked egg (7.8%) in UMI, and reduced-fat milk (7.4%) in HI were top food sources. Saimin noodles (13.7%) also contributed to folate in LMI jurisdictions, whereas enriched white rice contributed in both UMI (10.1%) and HI (7.5%) jurisdictions. Vitamin A, which is essential for vision and immune function, was primarily consumed through reduced-fat milk across WBIGs.

TABLE 5.

Top 3 food sources of vitamins and their percentage (%) contribution to daily intake by World Bank income group, the children’s health living study (2013–2014).

Vitamin	Rank	Lower middle income	%	Upper middle income	%	High income	%
Choline	1	Soup, saimin, from dry, oriental broth with noodles	5.4	Egg, hard-cooked	7.8	Milk, reduced fat (2%), fluid or reconstituted dry	7.4
	2	Spam	4.7	Milk, reduced fat (2%), fluid or reconstituted dry	5.5	Milk, low fat (1%), fluid or reconstituted dry	5.6
	3	Fish, dolphinfish (mahimahi), cooked	4.5	Egg, fried	5.3	Egg, scrambled	5.1
Folate (DFE)	1	Soup, saimin, from dry, oriental broth with noodles	13.7	Rice, white, enriched, short grain, cooked, no salt	10.1	Rice, white, enriched, short grain, cooked, no salt	7.5
	2	Rice, white, enriched, short grain, cooked, no salt	13.0	Cereal, cornflakes	9.0	Cereal, cheerios	6.9
	3	Doughnuts, yeast-type, plain	7.6	Bread, white, enriched	8.4	Bread, white, enriched	5.1
Niacin (B3)	1	Tuna, light meat, canned in oil, drained	10.2	Tuna, light meat, canned in oil, drained	7.4	Rice, white, enriched, short grain, cooked, no salt	3.3
	2	Rice, white, not enriched, short grain, cooked, no salt	9.3	Cereal, cornflakes	5.3	Bread, white, enriched	3.1
	3	Fish, dolphinfish (mahimahi), cooked	6.2	Bread, white, enriched	4.1	Rice, white, not enriched, short grain, cooked, no salt	3.0
Pantothenic acid (B5)	1	Rice, white, not enriched, short grain, cooked, no salt	39.7	Rice, white, not enriched, short grain, cooked, no salt	15.7	Rice, white, not enriched, short grain, cooked, no salt	12.9
	2	Milk, reduced fat (2%), fluid or reconstituted dry	3.5	Milk, reduced fat (2%), fluid or reconstituted dry	7.3	Milk, reduced fat (2%), fluid or reconstituted dry	9.5
	3	Fish, dolphinfish (mahimahi), cooked	3.0	Milk, low fat (1%), fluid or reconstituted dry	5.5	Milk, low fat (1%), fluid or reconstituted dry	6.4
Riboflavin (B2)	1	Milk, reduced fat (2%), fluid or reconstituted dry	8.1	Milk, reduced fat (2%), fluid or reconstituted dry	10.2	Milk, reduced fat (2%), fluid or reconstituted dry	12.2
	2	Rice, white, not enriched, short grain, cooked, no salt	6.8	Milk, low fat (1%), fluid or reconstituted dry	7.6	Milk, low fat (1%), fluid or reconstituted dry	8.1
	3	Soup, saimin, from dry, oriental broth with noodles	6.1	Cereal, cornflakes	6.8	Milk, whole, fluid	2.8
Thiamin (B1)	1	Soup, saimin, from dry, oriental broth with noodles	10.3	Cereal, cornflakes	7.8	Bread, white, enriched	4.4
	2	Rice, white, not enriched, short grain, cooked, no salt	9.7	Bread, white, enriched	7.2	Rice, white, enriched, short grain, cooked, no salt	4.2
	3	Rice, white, enriched, short grain, cooked, no salt	5.9	Soup, saimin, from dry, oriental broth with noodles	6.0	Milk, reduced fat (2%), fluid or reconstituted dry	3.4
	1	Milk, reduced fat (2%), fluid or reconstituted dry	9.7	Milk, reduced fat (2%), fluid or reconstituted dry	11.3	Milk, reduced fat (2%), fluid or reconstituted dry	12.8
Vitamin A (RAE)	2	Mackerel, canned, drained	6.0	Milk, low fat (1%), fluid or reconstituted dry	8.9	Milk, low fat (1%), fluid or reconstituted dry	9.1
	3	Fish, tuna, bluefin (ahi, maguro), raw	3.9	Cereal, cornflakes	4.4	Carrots, raw	4.2
Vitamin B6	1	Rice, white, not enriched, short grain, cooked, no salt	16.0	Cereal, cornflakes	9.8	Rice, white, not enriched, short grain, cooked, no salt	4.5
	2	Bananas, ripe	5.7	Bananas, ripe	6.1	Bananas, ripe	4.4
	3	Fish, dolphinfish (mahimahi), cooked	5.3	Rice, white, not enriched, short grain, cooked, no salt	5.7	Cereal, cornflakes	3.8
Vitamin B12	1	Mackerel, canned, drained	15.2	Mackerel, canned, drained	12.4	Milk, reduced fat (2%), fluid or reconstituted dry	11.5
	2	Sardines with bone, canned in oil, drained	14.1	Milk, reduced fat (2%), fluid or reconstituted dry	8.3	Milk, low fat (1%), fluid or reconstituted dry	7.3
	3	Tuna, light meat, canned in oil, drained	6.3	Cereal, cornflakes	8.0	Cereal, cornflakes	3.4
Vitamin C	1	Kool-Aid	11.7	Orange-apricot drink, canned (use for unspecified juice-type drink)	11.6	Oranges, raw, all varieties	9.9
	2	Orange-apricot drink, canned (use for unspecified juice-type drink)	7.3	Oranges, raw, all varieties	10.2	Orange drink, canned	8.6
	3	Breadfruit, boiled	7.2	Kool-Aid	8.8	Orange-apricot drink, canned (use for unspecified juice-type drink)	8.5
Vitamin D	1	Tuna, light meat, canned in oil, drained	20.9	Milk, reduced fat (2%), fluid or reconstituted dry	17.5	Milk, reduced fat (2%), fluid or reconstituted dry	26.2
	2	Mackerel, canned, drained	16.9	Milk, low fat (1%), fluid or reconstituted dry	15.7	Milk, low fat (1%), fluid or reconstituted dry	21.2
	3	Sardines with bone, canned in oil, drained	13.2	Tuna, light meat, canned in oil, drained	12.3	Milk, whole, fluid	5.7

Abbreviations: DFE, Dietary Folate Equivalent; RAE, retinol activity equivalent.

The primary food source of vitamin C was oranges for UMI and HI jurisdictions, and in LMI jurisdictions, sugar-sweetened beverages, like Kool-Aid, were commonly reported.

Mineral intake

Differences in mineral intake between WBIGs were evident, with calcium and potassium being the most concerning inadequacies observed; only ∼28% of children met dietary recommendations for both minerals (Table 2 and Supplementary Table 2). Calcium and potassium intakes were lower (below EAR) in LMI jurisdictions compared to UMI and HI jurisdictions. Calcium intake recommendations were met by only 5.6% of children in LMI jurisdictions compared to 37.0% and 39.5% of children in UMI and HI jurisdictions, respectively (Table 2 and Supplementary Table 2). Potassium intake met recommendations by 18.0% of children in LMI, 39.1% in UMI, and 28.1% in HI jurisdictions.

Although zinc intake recommendations were met by 91.1% of children in LMI jurisdictions, only 70.5% and 69.6% of children met recommendations in UMI and HI jurisdictions; however, these recommendations were unmet by exceeding the UL. A high proportion of children met the selenium intake recommendations, but the vast majority of those that did not meet recommendations exceeded the UL intake (LMI: 23.4%; UMI: 29.4%; HI: 12.9%). Recommended intakes for all other minerals (copper, iron, magnesium, manganese, phosphorus, and sodium) were met by over 90% of children across WBIGs.

Mineral intake and health outcomes

Overall, only selenium intake was inversely associated with OWOB status in children from the USAP region across all WBIGs (Table 3). Notably, 21.9% of children had selenium intake above tolerable UL (Table 2). Calcium and phosphorus intake were statistically lower (P < 0.05) among children with OWOB than among children without OWOB in LMI and HI jurisdictions. Additionally, in LMI jurisdictions, sodium intake was also lower among children with OWOB; however, the opposite was true in HI jurisdictions. High intake of iron and sodium were directly associated with OWOB in LMI and HI jurisdictions, respectively. In UMI jurisdictions, lower intakes of selenium and higher intakes of zinc were observed among children with OWOB compared to those without OWOB. Strikingly, zinc intake was not met by 22.9% of children participants in the study.

The presence of AN in children of the USAP region was associated with lower intake of several minerals, including calcium, copper, iron, phosphorus, and potassium (Table 4). When considering all participants, a positive association between the intake of selenium and the presence of AN was uncovered. After stratifying by WBIGs, this positive association was only found in HI jurisdictions (Table 3).

Stratification by WBIGs confirmed most of the associations between mineral intake and AN in HI jurisdictions, except for copper (Table 4). An inverse association between AN presence and zinc intake was additionally uncovered in the HI jurisdictions. In LMI jurisdictions, the presence of AN was associated with a lower intake of copper, magnesium, and potassium and a higher intake of iron and zinc. Strikingly, UMI jurisdictions had no association between mineral intake and AN presence (Table 4).

Top food sources of minerals

Notably, unenriched white rice was the primary source of magnesium, manganese, selenium, and zinc across LMI, UMI, and HI jurisdictions (Table 6). It also served as the top source of phosphorus, potassium, and iron in LMI jurisdictions, as well as copper in both LMI and HI jurisdictions. Despite some consistency in the top food sources of minerals, there were a few notable differences. Reduced-fat milk was the top source of calcium across LMI, UMI, and HI jurisdictions, contributing 14.3%, 19.1%, and 19.9% of total calcium intake, respectively. However, although low-fat milk and whole milk were the second and third top sources of calcium in UMI and HI jurisdictions, canned sardines (8.8%) and canned mackerel (7.7%) were the top 2 sources in LMI jurisdictions. For potassium, reduced-fat milk, low-fat milk, and ripe bananas were the top 3 sources in UMI and HI jurisdictions, whereas unenriched white rice, coconut water, and cooked mahimahi (dolphinfish) were the top sources in LMI jurisdictions.

TABLE 6.

Top 3 food sources of minerals and their percentage (%) contribution to daily intake by World Bank income group, the children’s health living study (2013–2014).

Mineral	Rank	Lower-middle income	%	Upper-middle income	%	High income	%
Calcium	1	Milk, reduced fat (2%), fluid or reconstituted dry	14.3	Milk, reduced fat (2%), fluid or reconstituted dry	19.1	Milk, reduced fat (2%), fluid or reconstituted dry	19.9
	2	Sardines with bone, canned in oil, drained	8.8	Milk, low fat (1%), fluid or reconstituted dry	14.4	Milk, low fat (1%), fluid or reconstituted dry	13.5
	3	Mackerel, canned, drained	7.7	Milk, whole, fluid	7.7	Milk, whole, fluid	4.5
Copper	1	Rice, white, not enriched, short grain, cooked, no salt	19	Papaya, ripe, raw	8.1	Rice, white, not enriched, short grain, cooked, no salt	5.8
	2	Water	5.8	Rice, white, not enriched, short grain, cooked, no salt	6.9	Water	4.2
	3	Taro, cooked	3.8	Bread, white, enriched	4.9	Bread, white, enriched	3.6
Iron	1	Rice, white, not enriched, short grain, cooked, no salt	9.6	Cereal, cornflakes	10.2	Cereal, cheerios	4.8
	2	Soup, saimin, from dry, oriental broth with noodles	6.6	Bread, white, enriched	5.7	Rice, white, enriched, short grain, cooked, no salt	4.1
	3	Rice, white, enriched, short grain, cooked, no salt	5.2	Rice, white, enriched, short grain, cooked, no salt	4.8	Cereal, cornflakes	4.0
Magnesium	1	Rice, white, not enriched, short grain, cooked, no salt	26.1	Rice, white, not enriched, short grain, cooked, no salt	11.1	Rice, white, not enriched, short grain, cooked, no salt	8.9
	2	Coconut water	5.5	Milk, reduced fat (2%), fluid or reconstituted dry	5	Milk, reduced fat (2%), fluid or reconstituted dry	6.4
	3	Fish, dolphinfish (mahimahi), cooked	2.7	Milk, low fat (1%), fluid or reconstituted dry	3.7	Milk, low fat (1%), fluid or reconstituted dry	4.3
Manganese	1	Rice, white, not enriched, short grain, cooked, no salt	57.3	Rice, white, not enriched, short grain, cooked, no salt	25.7	Rice, white, not enriched, short grain, cooked, no salt	21.0
	2	Rice, white, enriched, short grain, cooked, no salt	4.2	Tea, sweetened, canned, no vitamin c added	6.2	Rice, white, enriched, short grain, cooked, no salt	5.2
	3	Coconut milk, raw (liquid from grated meat and water)	3.1	Rice, white, enriched, short grain, cooked, no salt	5.9	Tea, sweetened, canned, no vitamin c added	3.5
Phosphorus	1	Rice, white, not enriched, short grain, cooked, no salt	18.3	Milk, reduced fat (2%), fluid or reconstituted dry	9.0	Milk, reduced fat (2%), fluid or reconstituted dry	11.1
	2	Tuna, light meat, canned in oil, drained	5.4	Milk, low fat (1%), fluid or reconstituted dry	6.8	Milk, low fat (1%), fluid or reconstituted dry	7.5
	3	Milk, reduced fat (2%), fluid or reconstituted dry	4.8	Rice, white, not enriched, short grain, cooked, no salt	6.7	Rice, white, not enriched, short grain, cooked, no salt	5.1
Potassium	1	Rice, white, not enriched, short grain, cooked, no salt	7.0	Milk, reduced fat (2%), fluid or reconstituted dry	7.4	Milk, reduced fat (2%), fluid or reconstituted dry	9.3
	2	Coconut water	6.6	Milk, low fat (1%), fluid or reconstituted dry	5.5	Milk, low fat (1%), fluid or reconstituted dry	6.2
	3	Fish, dolphinfish (mahimahi), cooked	4.6	Bananas, ripe	4.8	Bananas, ripe	3.5
Selenium	1	Rice, white, not enriched, short grain, cooked, no salt	24.0	Rice, white, not enriched, short grain, cooked, no salt	11.7	Rice, white, not enriched, short grain, cooked, no salt	10.9
	2	Tuna, light meat, canned in oil, drained	8.6	Tuna, light meat, canned in oil, drained	8.1	Milk, reduced fat (2%), fluid or reconstituted dry	3.1
	3	Turkey, roasted, skin only	5.5	Turkey, roasted, skin only	3.6	Milk, low fat (1%), fluid or reconstituted dry	2.7
Sodium	1	Sausage, frankfurter, beef and pork (unspecified)	10.7	Sausage, frankfurter, beef and pork (unspecified)	7.7	Sausage, frankfurter, beef and pork (unspecified)	5.3
	2	Spam	10.1	Bread, white, enriched	5.4	Spam	4.6
	3	Soup, saimin, from dry, oriental broth with noodles	8.0	Soup, saimin, from dry, oriental broth with noodles	5.4	Soy sauce, regular	3.7
Zinc	1	Rice, white, not enriched, short grain, cooked, no salt	25.2	Rice, white, not enriched, short grain, cooked, no salt	9.0	Rice, white, not enriched, short grain, cooked, no salt	6.1
	2	Sausage, frankfurter, beef and pork (unspecified)	5.1	Milk, reduced fat (2%), fluid or reconstituted dry	4.9	Milk, reduced fat (2%), fluid or reconstituted dry	5.3
	3	Spam	3.8	Sausage, frankfurter, beef and pork (unspecified)	3.7	Milk, low fat (1%), fluid or reconstituted dry	3.5

A striking source of sodium across all WBIGs was sausage and frankfurters (hotdogs), with other notable contributors including spam, saimin, and soy sauce. UMI showed a minor variation, with enriched white bread ranking as the second-largest sodium contributor. Coconut milk was a key source of manganese in LMI, whereas copper came primarily from grains and water. In LMI and UMI, starchy vegetables like cooked taro and fruits like papaya were also among the top sources of copper. Lastly, dolphinfish and tuna were particularly prominent sources of multiple vitamins and nutrients in LMI compared to UMI and HI jurisdictions.

Discussion

This study is the first to analyze the association between micronutrient intake and the prevalence of OWOB and AN in 2–8 y-old children from 11 jurisdictions of the USAP region. The results revealed inadequacies in the intake of calcium, potassium, sodium, niacin, vitamin D, and vitamin E across all jurisdictions, irrespective of income levels. These findings align with existing literature that highlights similar inadequacies in other Pacific-origin populations, such as Kiribati, Alaska, and New Zealand [40,[58], [59], [60]]. In contrast, school children aged 9–12 y-old from Guam, a USAP jurisdiction, had insufficient intake of calcium, vitamin E, and folate while consuming excessive amounts of iron and zinc [61]. However, the age group analyzed varies in each study, and thus, these findings are not fully comparable with the results obtained in the present study.

Calcium, potassium, and vitamin D have all been identified as nutrients of concern due to underconsumption among the general United States population [62] and children in the Pacific region [40,[58], [59], [60]]. The findings presented here uphold that this concern exists regardless of WBIG. Milk, a school meal choice, was identified as the top source of calcium, potassium, and vitamin D, and yet recommendations are not being met [63], suggesting that consumed amounts may either need to be increased or alternative food sources, such as small fish with edible bones, should be added to the children’s diets. Deficiencies in these micronutrients lead to an increased risk of rickets and detrimental bone development in children, even when their daily intake of phosphorus is sufficient. Although widespread in the USAP, calcium and potassium inadequacies were higher in the FSM, an LMI area, than in other jurisdictions.

Furthermore, vitamin D synthesis in the body requires sun exposure to convert its precursors into the active hormone calcitriol, and aside from Alaska, the United States Pacific jurisdictions are located in areas with high levels of sun exposure throughout the year. Since vitamin D regulates phosphorus and calcium balance if low dietary intake is not compensated with sun exposure, vitamin D deficiency could lead to hyperphosphatemia. Follow-up with biomarker status would be valuable. In adults, hyperphosphatemia compromises kidney function and blood pressure control and is a significant predictor of cardiovascular mortality [64]. Currently, it is unclear if hyperphosphataemia accompanied by vitamin D and calcium deficiencies in children triggers the same metabolic outcome as in adults. Dietary diversity can ensure adequate nutrition and decrease the reliance on specific foods that may not be sustainable or feasible in the children’s food environment. For example, milk is generally a rich source of nutrients, but the majority of jurisdictions that participated in this study rely on importing milk, an economic policy necessary in the absence of local dairy farms.

Micronutrient deficiencies are often a major concern for children’s growth and metabolic health. However, this study also identified unmet micronutrient needs due to excessive amounts in children’s diets, which are also detrimental to metabolic health and can drive obesity and insulin resistance. Widespread excessive intake of folate and niacin was found regardless of WBIG, raising concerns of toxicity. Enriched and fortified foods such as milk, cereal, and white bread were commonly reported as sources of these micronutrients and may more easily provide excess nutrients compared to fruit and vegetables, which were not sufficiently present in children’s diets [63]. Excess folate was particularly prevalent in UMI and HI areas (Supplemental Table 2). Although folate UL quantities are still being debated [65], the levels observed in children of the UMI and HI areas of USAP can be considered troublesome. Folate is commonly supplemented in fortified foods such as flour and rice, which are primary food staples in the Pacific, raising safety concerns for these populations. Folate is essential to 1-carbon metabolism, supporting the methionine cycle and transsulfuration pathway in cells, which are involved in the epigenetic control of genes for metabolic enzymes and oxidative stress defense, respectively [66,67]. Excess folate could disturb redox balance in metabolic tissues, enhancing protein oxidation with dire consequences to energy homeostasis. Excess folate also leads to neuropathy in vitamin B12-deficient adults [68], and negative metabolic outcomes of folate excess combined with vitamin B12 deficiency in children have been reported [69]. Similarly, excess niacin can promote cardiovascular diseases in adults by increasing inflammation [70]. Although the effects of excess niacin in children are not yet well understood, they should not be overlooked. The intake of niacin among USAP children was predominantly excessive and, if found to be harmful, this could indicate that USAP children are at higher risk for these metabolic conditions, a significant consideration for future policies and nutritional guidelines.

This study also uncovered an alarmingly high intake of sodium in the children of the USAP region (Table 2 and Supplementary Table 2). High sodium is the main driver of cardiovascular diseases in adult populations, and high intake of sodium in the first 6 mo of life impacts cardiovascular outcomes later on [71]. Thus, this result may point to future impacts on cardiovascular health, which, combined with the strong association with obesity and sodium concentrations, leads to a need for the reduction of this micronutrient as a main policy target in children’s dietary health [72]. Campaigns to reduce sodium intake by targeting restaurants and other food establishments have been deployed in the region [73], yet concrete improvements in cardiovascular health for the population are still undetermined. The current study shows that the main sources of sodium in children’s diets include sausage, canned meats, and instant noodles. This prevalence could be a result of the rapid nutrition transition to a Westernized diet in the region, as these goods are inexpensive, commonly imported, and easy to store. Thus, a likely add-on to a campaign aiming to reduce sodium intake in children could be employing strategies to make alternative sources of meat that are less processed more available and cheaper.

The data also unveiled that iron intake was mostly adequate among USAP children, even when separated by WBIG. This is an improvement compared to previous data from American Samoa children, which revealed alarming rates of anemia, iron deficiency, and iron deficiency anemia in this population [74]. Top sources of iron across jurisdictions are predominantly from fortified grains—i.e., enriched rice in LMI and cereal in UMI and HI. It is possible that iron deficiency in children was recognized as a concern by policymakers, with nutritional and infectious disease policy changes favoring fortified products in school meals that ultimately corrected this deficiency in the time span between both findings. However, it should be hallmarked that the methodology of this study only analyzed iron intake, whereas the study in American Samoa analyzed iron status in hemoglobin and heme concentrations. This raises the additional possibility of the methodological discrepancy between the 2 studies as a source for the contrasting results, i.e., an AI occurs, but it does not sustain sufficient hemoglobin production, and children can still be anemic due to infectious genetic and metabolic differences in the child populations. Although unlikely, this possibility should still be cautiously considered.

The mixed-income nature of the USAP region was resolved in this study by stratification of the results according to WBIG. This approach provides a unique opportunity to uncover a refined picture of micronutrient intake and its association with obesity and AN in children that is relevant to dietary guidelines in the Pacific. For example, children from HI jurisdictions of the USAP consumed more processed and fortified food products and had higher estimates of OWOB. This suggests a strategy is needed to either decrease consumption of processed and fortified products in HI jurisdictions and/or better regulate the fortification process by including micronutrients that are deficient, such as vitamin D, and reduce amounts of the ones that are in excess, such as folate. Still, 43% of adults in the Pacific region are OWOB, according to WHO health data, and the prevalence of children with OWOB varies from single digits to almost 50%, depending on jurisdictions.

Micronutrient malnutrition is prevalent in obesity [75], leading to the “hidden hunger” phenomena that can precede the development of major metabolic problems such as cardiovascular diseases, stroke, and type 2 diabetes [22,76]. This occurs because most micronutrients participate in key molecular pathways that control energy homeostasis, lipid metabolism, and adipose tissue physiology. For example, selenium is an essential component of selenoproteins, which are mostly involved in curbing oxidative stress in adipocytes, pancreatic β-cells, and liver, all key players in the regulation of glucose and lipid homeostasis [[77], [78], [79]]. Acting on a U-shape curve pattern, selenium’s detrimental effects can occur either in deficiency or in excess. In this study, we found that childhood OWOB was associated with lower selenium intake and higher thiamin intake. Although obesity in adults is typically linked to higher selenium intake in areas with selenium-rich soils, such as the Pacific region [80], in this study, children with OWOB in the USAP region had lower selenium intake. About 22% of the children exceeded the UL for selenium intake across all jurisdictions. However, even the modest reduction in selenium intake predicted an increased risk of OWOB, suggesting a need for precision nutrition strategies specifically focused on selenium intake, especially in UMI jurisdictions. These findings align with a large meta-analysis that found no direct link between obesity and selenium status but indicated that individuals with higher adiposity metabolize selenium differently, resulting in lower selenium concentrations in red blood cells and impairment of key selenoproteins that protect against oxidative stress [80]. Although the meta-analysis focused on adults, it is possible that a similar metabolic disruption occurs in children with obesity, which could contribute to the pathophysiology of obesity.

The association between OWOB and high thiamin intake in USAP children is puzzling. It is common for adults with obesity to have thiamin deficiency [81]. Thiamine is an essential cofactor to enzymes involved in pyruvate metabolism and the tricarboxylic acid cycle, both fundamental pathways to carbohydrate, lipid, and protein metabolism. Having thiamine detrimental effects on substrate metabolism behave in a U-shaped curve suggests that small changes in intake are sufficient for obesity-related pathological outcomes. The thiamine intake level in these children may be a marker for the high rice intake, thus far without have yet metabolic consequences.

In LMI jurisdictions, OWOB was associated with an inadequate intake of vitamin A, calcium, iron, phosphorus, and sodium in children. These nutrients are primarily found in reduced-fat milk, rice, and processed meats, which are predominant in the dietary pattern of USAP. In contrast, in HI jurisdictions, the micronutrient profile associated with OWOB showed inverse associations with pantothenic acid, calcium, and phosphorus intake. Both groups of children consumed similar foods as the primary sources of these micronutrients. This suggests a common starting point for improving children’s diets in HI and LMI jurisdictions. Strategies such as biofortification or enriching these food items with slightly higher amounts of these nutrients—achievable, for example, by changing vendor sources —could help address these dietary inadequacies. In UMI jurisdictions, OWOB was associated with higher zinc intake. Almost 2 billion people globally are zinc deficient regardless of their socio-economic status [82,83]. In children, zinc deficiency causes stunting and cognitive impairment, an effect that hinges on zinc’s integral binding to ∼10% of the human proteome in cysteine residues, impacting basic cellular functions such as cell cycle progression and DNA synthesis and consequently development [84]. Moreover, high zinc stimulates adipocyte lipogenesis even without insulin [85], and this early effect can be foundational to the consequences of excess intake and its association with OWOB in children. Excessive zinc intake leads to reduced copper concentrations, as zinc enhances copper sequestration in metallothioneins in enterocytes and excretion [86], triggering copper deficiency. Excess zinc can also lead to reduced HDL cholesterol and compromised immunity [87]. These effects may increase the risk of infections and dyslipidemia in children with OWOB. The primary sources of zinc in UMI jurisdictions were unenriched white rice, with canned tuna and milk as secondary sources. Since milk is an important source of calcium, which is often deficient in Pacific children, increasing milk consumption to address calcium deficiency should be carefully considered.

The study also uncovered the association of micronutrient intake with AN, a dermatological hyperpigmentation that serves as a proxy for insulin resistance [49,88]. The presence of AN in children was associated with inadequate intake of folate, thiamin, vitamin E, iron, potassium, and zinc in LMI and HI jurisdictions. Molecular mechanisms linking micronutrients directly with AN development are unknown and likely hinge on their effects on insulin signaling. For example, zinc is considered an insulin sensitizer [89], whereas high concentrations of iron decrease insulin signaling phosphorylation in adipocytes [90]. However, more studies are needed to determine additional pathways triggered by insulin resistance that lead to the skin phenotype of AN. In children from UMI jurisdictions, there was no association between AN presence and micronutrient intake, except for choline. The intake of choline in children from all WBIG jurisdictions was lower than the recommended levels, raising concerns about choline nutrition in the region, yet higher choline intake was observed in children with AN in UMI jurisdictions. Choline has beneficial effects on the neurodevelopment of children, and observational studies have positively associated choline with child growth and neurocognition in low-income countries [91,92] and among United States children affected by food insecurity [93]. This discrepancy suggests that the choline intake required for metabolic health and neurodevelopment in children differ, and more refined dietary guidelines, balancing the obesity burden and the children’s healthy growth according to the population’s WBIG, are needed. Moreover, the discrepancies between the association of micronutrients with AN in the different jurisdictions point to AN as a complex condition in need of a basic understanding of its connection with micronutrients and nutritional interventions involving micronutrients.

The results show that micronutrient intake imbalances are valuable associative markers of OWOB and insulin resistance on the extremes of income groups in the USAP region. These findings should inform future strategies for correcting micronutrient inadequacies in the Pacific according to the economic status of the jurisdiction. In particular, it should focus on LMI and HI, as these groups are significantly more prone to display hidden hunger than UMI jurisdictions. These associations, in combination with the identification of the most consumed source items in the children’s diet for each micronutrient, point to specific targets for food biofortification and enrichment, nutritional education, and dietary policies that are relevant according to regional dietary patterns and food access to improve the health of the USAP children.

Recommendations

• -Reduction of sodium in dietary intake is a main policy target for USAP children.
• -Continue to enhance policies on providing children with iron-fortified grains in school meals.
• -The fortification of products for children, including vitamin D and calcium, and reducing folate, especially for LMI and HI jurisdictions.
• -Focused nutritional strategies to improve micronutrient intake should focus on LMI and HI jurisdictions.

Limitations

A major limitation of the results is the absence of measurements for plasma concentrations of the micronutrients of interest. The correspondence between micronutrient intake and levels in circulation may not be straightforward, as genetics, micronutrient metabolism, digestive and excretory mechanisms, and microbiome composition contribute to interpersonal variance that could lead to discrepancies. For example, it is challenging to infer the selenium content of seafood and plant-derived items, depending on the biology of the animal or plant species being consumed, as well as levels in different jurisdictions, especially in the context of local farming. Moreover, individual variability in uptake and excretion mechanisms, known to be an intrinsic characteristic of selenium metabolism in humans [94], may widen the discrepancy between intake and status. Nevertheless, assessment of food intake is essential to guide food policy changes and improve nutritional health for a population.

A second limitation of this study was the available food composition data. The analysis utilized the latest PacTrac food composition data to calculate micronutrient intakes [52]. However, as with all dietary assessment tools, there is a reliance on default food items for food entries that lack detailed food composition data, such as “fish” being recorded as “dolphinfish,” which may not provide a true representation of the cultural and regional diversity and specificity of the food items consumed. Databases do not have complete local foods or environmental variability of those foods by jurisdiction.

Third, the WBIG classification served here as a proxy for food accessibility and economic resources.

A fourth limitation of this study is that it did not include information on where the foods were eaten by the children, which was collected from the diet records. School policy is a critical determinant of food choices for children; however, one cannot infer from this study whether policies to improve micronutrient deficiencies and reduce toxicities could be directed to schools or better turned into family education on household diets. Although it is likely both components could be targeted, a precise approach focusing on specific micronutrients or vitamin inadequacy may suggest 1 location more than another.

In conclusion, this study profiles multiple micronutrient intake in young children across the USAP region and examines its association with metabolic conditions, considering the jurisdiction income group where these children reside. AI of dietary micronutrients is critical for children’s growth and development, yet the findings reveal that many children in the USAP region do not meet the recommended intakes for calcium, potassium, niacin, vitamin D, and vitamin E. The primary food sources contributing to micronutrient intake provide important insights into the dietary patterns of children across different income groups, offering a basis for tailored, cost-effective nutrition education and intervention strategies. A comprehensive, long-term micronutrient profile pattern among health outcomes could help detect hidden hunger and inadequacy trends, helping provide precision nutrition guidance for the region. This research, along with previous studies in other Pacific Island communities, is a step to refine future nutritional guidelines that are relevant and specific to the needs of USAP populations.

Author contributions

The authors’ responsibilities were as follows– RN, LAS, ABY, LRW: concept and design where the authors had full access to the data in this study and are responsible for the integrity and the accuracy of data analysis; RN, LRW: data acquisition where the authors were responsible for the Children’s Healthy Living Center of Excellence administration, which encompassed TFA, PC, TF, JD, and LS; LAS, ABY, RN, KH, EL; data analysis and interpretation; LAS, ABY, KH: writing of the manuscript; all authors performed critical revision of the manuscript for intellectual content; ABY, KH, EL: statistical analysis; RN, LAS, LRW: obtained funding; TA, PC, TF, JD, LS: administrative, technical, and material support; RN: supervision; and all authors: read and approved the final manuscript.

Data availability

Data described in the manuscript and data dictionary will be made available upon request pending approval by the Children’s Healthy Living Program steering committee, agreement to data security requirements, and payment of processing fees.

Funding

Funding to support research efforts and personnel in this study was provided by the USDA, Agriculture and Food Research Institute Initiative (grants 2011-68001-30335, 2018-69001-27551, and 2021-68012-35899) from the National Institute of Food and Agriculture; National Institutes of General Medical Sciences (grant P20GM139753 for the Statistical Analysis of Nutrition and Diet Core of the Integrative Center for Precision Nutrition and Human Health to RN and ABY); National Institute of Diabetes and Digestive and Kidney Diseases (grant R01DK128390 to LAS); National Institute on Minority Health and Health Disparities (grant U54MD007601 to EL); Ingeborg v.F. McKee Fund of the Hawaii Community Foundation (grant MedRes_2023_00002973 to LAS). The design of the study, collection, analyses, or interpretation of data, writing of the manuscript, and the decision to publish the results are solely the responsibility of the authors and do not necessarily represent the official views of any of the funding agencies that supported this investigation.

Conflict of interest

LRW is an editorial board member for the current developments in nutrition and played no role in the journal’s evaluation of the manuscript.

Appendix A. Supplementary data

The following is the Supplementary data to this article: