- Email: [email protected]
Dietary Intake in Head Start vs Non–Head Start Preschool-Aged Children: Results from the 1999-2004 National Health and Nutrition Examination Survey
RESEARCH Original Research
Dietary Intake in Head Start vs Non–Head Start Preschool-Aged Children: Results from the 1999-2004 National Health and Nutrition Examination Survey EMILY M. BUCHOLZ, MPH; MAYUR M. DESAI, PhD, MPH; MARJORIE S. ROSENTHAL, MD, MPH
ABSTRACT Objective To determine whether dietary intakes of children enrolled in Head Start programs differ from those of children not attending preschool or children in non–Head Start programs. Design Using data from the 1999-2004 National Health and Nutrition Examination Survey, low-income, 3- to 5-year-old children were categorized into one of four preschool groups: Head Start (n⫽184), non–Head Start (n⫽189), past preschool (n⫽193), and no preschool (n⫽384). Total nutrient intakes were calculated using 24-hour parental recalls. Statistical analyses performed Mean macronutrient and micronutrient intakes were compared across groups and the percentage of children not meeting Recommended Dietary Allowances (RDAs) were calculated. Multivariate logistic regression was used to evaluate the relationship between preschool group and likelihood of not meeting dietary guidelines. Results Many children did not meet the RDA for folate (20.5%), vitamin A (39.7%), vitamin E (79.7%), calcium (40.2%), iron (28.8%), and potassium (90.8%). Compared with the other preschool groups, Head Start children had lower mean protein, saturated fat, riboflavin, calcium, and phosphorous intakes. The greatest differences in intake were observed between Head Start participants and no-preschool children. Multivariate analyses demonstrated an association between Head Start and inadequate intake of protein, thiamin, riboflavin, niacin, calcium, and selenium.
E. M. Bucholz is a medical student, M. M. Desai is an assistant professor, Division of Chronic Disease Epidemiology, and M. S. Rosenthal is an associate research scientist, Department of Pediatrics, and assistant director, Robert Wood Johnson Foundation Clinical Scholars Program, Yale School of Medicine, New Haven, CT. Address correspondence to: Emily M. Bucholz, MPH, Yale School of Medicine, 333 Cedar St, New Haven, CT 06511. E-mail: [email protected] Manuscript accepted: January 24, 2011. Copyright © 2011 by the American Dietetic Association. 0002-8223/$36.00 doi: 10.1016/j.jada.2011.04.009
© 2011 by the American Dietetic Association
Conclusions Compared with other low-income children, those in Head Start programs appear to be at greater risk for not meeting the RDA for several key vitamins and minerals. These differences in diet quality may present an opportunity for Head Start programs to enhance nutrition in this student population. J Am Diet Assoc. 2011;111:1021-1030.
P
reschool-aged children are at risk for inadequate nutrition and, among low-income preschoolers, the risk is even greater: 30% of preschool-aged children living below the poverty line are overweight or at risk of overweight (1) compared with 26% of all 2- to 5-year-olds (2). Children living in poverty are also at higher risk of micronutrient deficiencies, particularly zinc and iron (35). Given that the preschool years are when children begin exercising greater control over what they eat and establish eating patterns that may last a lifetime (6-9), the environmental context of a preschool-aged child may play a key role in lifetime nutrition. Nationwide, 60% of all US children receive at least some child care outside of the home, and one half of these children spend 35 hours or more each week in out-ofhome child care (10). As a result, many children consume more than one half of their daily food intake in out-of-home care. Head Start (HS), a federally funded preschool program serving low-income children, aims to “promote school readiness by enhancing the social and cognitive development of children through the provision of educational, health, nutritional, social and other services to enrolled children and families” (11). This program may serve as an opportunity to enhance nutritional intake. All HS programs are required to comply with federal Program Performance Standards that address the identification of children’s nutrition needs, menu planning and dietary requirements for meals served in the classroom, feeding practices and nutrition education, food safety and sanitation, and family education around child nutrition (12). Despite risk of inadequate nutrition for low-income preschool-aged children and the emphasis HS places on health, there is a paucity of literature on the nutritional benefits of HS participation. The purpose of this study was to examine, using a nationally representative sample of low-income children, whether dietary intakes of children enrolled in HS programs differ from those of chil-
Journal of the AMERICAN DIETETIC ASSOCIATION
1021
dren not attending preschool or children in non-HS preschool programs. METHODS Data Source Data from the National Health and Nutrition Examination Survey (NHANES) were used to assess dietary intake in US preschool-aged children. Conducted by the National Center for Health Statistics, NHANES is a nationally representative, cross-sectional survey of the civilian, noninstitutionalized population of the United States. It uses stratified, multistage probability sampling to oversample low-income individuals, certain age groups, and races/ethnicities (13). For this study, data from three NHANES cohorts was combined: 1999-2000, 2001-2002, and 2003-2004. NHANES collects detailed demographic, socioeconomic, dietary, and health-related data through in-home interviews and clinical assessments at a mobile examination center. For children younger than age 6 years, these interviews were conducted using parent or guardian proxies. A detailed description of these methods is available elsewhere (11,13). Selection of Participants and Measures The sample consisted of all children aged 3 to 5 years living in poverty (200% of the federal poverty line and below) who had available dietary intake data (N⫽950). We limited the sample to children whose annual family income fell at or below 200% of the federal poverty line for household size and year to compare children of similar socioeconomic status while still including the majority of children in HS programs. Preschool Status Participants were categorized into one of four groups according to history of preschool attendance using the following questions: “Is sample person [SP] now attending HS?” “Does SP now attend day care or preschool?” and “Did SP ever attend day care or preschool?” The four categories used were HS (children currently in HS), non-HS preschool (children in a non-HS preschool/child care program), past preschool (children who previously attended but no longer attend a preschool/child care program), and no preschool (children who have never attended a preschool/child care program). Dietary Intake and Serum Micronutrient Levels The NHANES total nutrient databases were used to estimate nutrient intakes for each group of children. Information on individual foods consumed by participants was collected by trained NHANES personnel using 24-hour parental recalls and then converted to total nutrient intakes using the US Department of Agriculture’s Food and Nutrient Database for Dietary Studies (14). Before 2003, a single dietary interview was administered during the mobile examination center exam to assess individual food and total nutrient consumption on 1 day only. Since then, NHANES has added an additional 24-hour recall to collect data on 2 days of intake, using
1022
July 2011 Volume 111 Number 7
both in-person and telephone interviews. Due to differences in the number of days for which dietary information was available for children in this study, analyses were restricted to the 1-day in-person data collected in the mobile examination center. To estimate the prevalence of children not meeting the recommended intake levels for specific macro- and micronutrients, total nutrient intakes were compared to age-specific standards. The standards used were Dietary Reference Intakes for protein and all reported micronutrients as defined by the Institute of Medicine’s Food and Nutrition Board guidelines (15). Because lowincome children are already at high risk of nutrient deficiencies, the more stringent criteria Recommended Daily Allowances (RDAs) were employed, as opposed to Estimated Average Requirements. RDAs are calculated to meet the need of 97% to 98% of individuals whereas Estimated Average Requirements only meet the needs of 50% of individuals. When RDAs were not available, Adequate Intakes were used. In the case of sodium, the tolerable Upper Level was used as a cut-off because of concerns over excess sodium. Guidelines for fat intakes were defined as not exceeding 35% to 40% of total energy from fat and 10% from saturated fat based on the US Department of Agriculture’s Dietary Guidelines for Americans 2005 (16). When available, serum concentrations of micronutrients were examined to determine whether differences in reported dietary intakes between preschool groups could be explained by differential misreporting. Methods describing the measurement of these variables and their associated errors are published elsewhere (13). Because NHANES collected different serum micronutrients in each of the three cohorts, the sample size and number of cohorts used for the analysis of each vitamin or mineral are indicated. Covariates The association between preschool group and several demographic, socioeconomic, and health-related variables was examined. These variables included child sex, age, race/ethnicity, insurance status, birthplace, and body mass index classification (overweight: ⬎95th percentile, at risk for overweight: ⬎85th to ⬍95th percentile, or normal: ⬍85th percentile) (17). Consistent with NHANES data collection, race/ethnicity was categorized as nonHispanic white, non-Hispanic black, Mexican American, and other. Socioeconomic variables included household size, presence of adult smoker(s) in the household, household poverty level, household Special Supplemental Nutrition Program for Women, Infants, and Children and/or food stamp receipt, the education level and birthplace of the household reference person, whether the child’s mother was a teenager while pregnant, smoked during pregnancy, or breastfed the child as an infant, and the child’s use of dietary supplements. Finally, to determine the possibility of differential recall based on day of the week, a dummy variable indicating weekend vs weekday recall was examined. Covariates associated with preschool enrollment (at the P⬍0.10 level) were included in the final multivariate models.
Table 1. Characteristics of US low-income preschool-aged children aged 3 to 5 years in the 1999-2004 National Health and Nutrition Examination Survey by preschool status (N⫽950)a Preschool Status
Variable
Overall
Head Start (unweighted nⴝ184)
Non–Head Start preschool (unweighted nⴝ189)
Past preschool (unweighted nⴝ193)
No preschool (unweighted nⴝ384)
P valueb
4™™™™™™™™™™™™™™™™™™ weighted %⫾standard error ™™™™™™™™™™™™™™™™™™™3 Child variables Girl Age (y) 3 4 5 Race/ethnicity Non-Hispanic white Non-Hispanic black Mexican American other Child born outside US Medical insurance Private Medicaid/SCHIPc Other insurance Uninsured Household variables Below federal poverty line Household size (including child) 2 3-4 5 or more Presence of adult smoker Special Supplemental Nutrition Program for Women, Infants, and Children benefits received Food Stamps received Household reference person born outside US Household reference person education level Less than high school High school Some college College graduate Pregnancy variables Teenage pregnancy Mother smoked while pregnant Child breastfed as infant Child health status variables Body mass index Normal (⬍85%) At risk of overweight Overweight (⬎95%) Supplement use Other variables Recall day of the week Weekend Weekday
51.0⫾1.9
52.7⫾4.4
46.7⫾5.4
56.0⫾4.9
49.7⫾3.5
0.568 ⬍0.001
35.8⫾2.1 32.4⫾2.3 31.8⫾1.8
29.1⫾5.5 40.3⫾4.6 30.5⫾5.2
28.2⫾3.6 42. ⫾4.7 29.7⫾5.3
25.0⫾4.1 26.4⫾4.2 48.5⫾4.5
48.7⫾3.5 27.6⫾4.5 23.7⫾3.1
48.2⫾3.9 18.6⫾2.2 19.1⫾2.5 14.2⫾2.4 3.3⫾0.7
33.6⫾6.6 37.3⫾5.7 20.3⫾4.2 8.8⫾3.4 3.6⫾1.2
52.5⫾5.6 21.2⫾2.8 11.8⫾2.5 14.6⫾4.0 2.8⫾1.2
53.4⫾7.0 20.2⫾4.3 11.6⫾2.7 14.8⫾4.3 2.5⫾1.7
48.2⫾5.7 9.2⫾1.8 26.9⫾3.8 15.6⫾3.2 4.0⫾1.1
30.9⫾2.3 47.8⫾2.5 8.2⫾1.2 13.1⫾1.8
19.8⫾3.6 57.0⫾4.2 8.7⫾2.7 14.6⫾3.3
35.3⫾5.2 47.5⫾5.4 9.1⫾3.4 8.1⫾2.5
37.3⫾5.1 41.6⫾5.0 7.0⫾2.5 14.0⫾3.9
28.9⫾3.0 48.3⫾4.2 8.1⫾1.6 14.7⫾2.8
50.9⫾2.0
61.6⫾4.4
40.0⫾4.9
45.2⫾4.7
56.1⫾3.3
5.3⫾0.9 47.3⫾2.8 47.4⫾3.0 31.9⫾3.4
5.4⫾2.1 48.9⫾5.1 45.7⫾4.8 30.3⫾5.1
8.2⫾2.8 55.4⫾4.8 36.4⫾5.1 20.8⫾4.7
6.7⫾2.0 58.8⫾5.4 34.5⫾4.8 33.1⫾5.1
2.9⫾1.2 35.6⫾3.5 61.5⫾3.8 37.9⫾5.3
0.040
46.3⫾2.7 66.5⫾3.6
55.6⫾4.9 72.6⫾5.3
38.6⫾5.4 65.0⫾7.6
41.5⫾3.8 65.6⫾6.7
49.9⫾4.4 65.4⫾5.6
0.049 0.860
24.2⫾2.6
32.3⫾3.9
17.2⫾3.9
17.7⫾3.7
22.7⫾3.6
⬍0.001 0.004
10.3⫾1.2 57.0⫾2.9 24.9⫾2.8 7.9⫾1.8
9.7⫾2.5 62.1⫾6.6 26.6⫾6.7 1.6⫾0.8
4.4⫾1.3 55.1⫾5.0 32.7⫾4.4 8.0⫾2.8
5.3⫾1.8 64.1⫾5.5 23.3⫾4.1 7.3⫾2.6
16.9⫾3.1 51.8⫾3.9 20.7⫾5.2 10.7⫾3.4
20.0⫾1.8 26.2⫾3.5 50.8⫾3.1
28.9⫾4.2 19.0⫾5.5 44.1⫾4.3
23.7⫾4.5 24.8⫾5.4 50.0⫾6.0
18.3⫾3.0 24.5⫾3.3 53.2⫾5.9
15.8⫾2.6 30.6⫾6.4 52.4⫾4.6
70.6⫾2.4 12.5⫾1.4 16.9⫾1.9 34.8⫾2.6
68.9⫾5.5 14.5⫾3.2 16.6⫾4.5 22.2⫾3.7
69.9⫾4.8 11.9⫾3.7 18.2⫾4.2 40.9⫾4.7
68.0⫾5.5 12.8⫾2.4 19.2⫾4.0 40.9⫾4.2
73.1⫾3.1 12.0⫾2.6 14.9⫾2.6 32.5⫾4.7
30.4⫾2.1 69.6⫾2.1
29.5⫾3.4 70.5⫾3.4
35.1⫾4.4 64.9⫾4.4
25.2⫾4.3 75.8⫾4.3
31.3⫾3.8 68.7⫾3.8
⬍0.001
0.844 0.267
0.005 ⬍0.001
0.043 0.430 0.661 0.963
0.026 0.403
a
Percentages may not sum to 100% because of rounding. Rao-Scott 2 tests were used to compare the distribution of characteristics across all four groups. c SCHIP⫽State Children’s Health Insurance Program. b
July 2011 ● Journal of the AMERICAN DIETETIC ASSOCIATION
1023
Table 2. Mean macro- and micronutrient intake reported by low-income, 3- to 5-year-old children in the 1999-2004 National Health and Nutrition Examination Survey by preschool status Preschool Statusa
Overall
Intake
Meanⴞstandard error
Non–Head Start preschool (unweighted nⴝ189)
Head Start (unweighted nⴝ184)
Past preschool (unweighted nⴝ193)
No preschool (unweighted nⴝ384)
4™™™™™™™™™™™™™™™™™™ weighted mean⫾standard error ™™™™™™™™™™™™™™™™™™3 Macronutrient Total energy (kcal) Total protein (g) Total carbohydrates (g) Total fat (g) Saturated fat (g) Cholesterol (mg) Fiber (g) Micronutrient Vitamin A (RAEb) (g) Vitamin B-1 (thiamin) (mg) Vitamin B-2 (riboflavin) (mg) Vitamin B-3 (niacin) (mg) Vitamin B-6 (mg) Vitamin B-9 (folate) (g) Vitamin B-12 (g) Vitamin C (mg) Vitamin E (mg) Calcium (mg) Phosphorous (mg) Magnesium (mg) Iron (mg) Zinc (mg) Copper (mg) Sodium (mg) Potassium (mg) Selenium (mg)
1,720.05⫾33.67 58.65⫾1.18 234.88⫾4.81 63.20⫾1.71 23.24⫾0.70 207.53⫾7.61 11.14⫾0.27
1,668.74⫾58.93 55.51⫾2.37 233.01⫾8.64 59.91⫾2.34 21.28⫾0.89 211.81⫾15.74 11.72⫾0.58
1,710.14⫾56.86 57.54⫾2.23 235.68⫾8.50 62.41⫾2.56 22.62⫾1.09 194.16⫾16.86 10.89⫾0.60
1,657.98⫾58.80 56.13⫾2.39 229.95⫾9.34 59.54⫾2.32 21.83⫾0.76 175.64⫾11.56 10.42⫾0.69
1,780.89⫾60.98 61.89⫾2.03* 238.02⫾8.75 67.00⫾3.30 25.14⫾1.34* 231.84⫾13.18 11.48⫾0.42
501.01⫾13.86 1.38⫾0.03 1.98⫾0.04 14.59⫾0.32 1.51⫾0.03 333.28⫾9.24 4.26⫾0.15 100.78⫾3.94 4.76⫾0.13 911.06⫾25.83 1,096.62⫾23.15 202.59⫾4.23 13.12⫾0.27 9.45⫾0.23 0.88⫾0.02 2,629.50⫾51.08 2,154.04⫾53.52 74.53⫾1.33
468.78⫾22.59 1.33⫾0.06 1.85⫾0.07 15.95⫾0.78 1.49⫾0.07 321.18⫾15.67 4.37⫾0.63 109.45⫾10.73 4.92⫾0.20 785.85⫾43.05 1,010.72⫾42.91 201.07⫾7.30 13.08⫾0.53 9.01⫾0.42 0.87⫾0.03 2,620.06⫾103.40 2,144.78⫾80.76 72.14⫾3.59
480.04⫾29.78 1.34⫾0.05 1.88⫾0.08 16.65⫾0.56 1.46⫾0.07 318.76⫾17.44 4.16⫾0.23 96.51⫾8.68 5.04⫾0.28 875.53⫾54.23 1,088.26⫾37.55 201.62⫾9.19 13.11⫾0.55 9.05⫾0.44 0.85⫾0.04 2,536.63⫾116.91 2,085.32⫾112.95 70.50⫾2.61
501.87⫾30.60 1.41⫾0.07 1.96⫾0.09 16.87⫾0.69 1.50⫾0.09 318.16⫾17.46 4.18⫾0.27 88.65⫾6.90 4.09⫾0.17** 862.16⫾36.49 1,042.47⫾40.83 189.46⫾8.08 12.59⫾0.61 9.17⫾0.49 0.82⫾0.04 2,607.23⫾82.42 2,006.08⫾95.08 70.55⫾2.47
523.97⫾27.24 1.41⫾0.04 2.08⫾0.08* 16.39⫾0.57 1.55⫾0.06 354⫾19.13 4.58⫾0.21 106.87⫾8.30 4.95⫾0.21 1,006.27⫾49.89** 1,165.31⫾39.97** 211.4⫾7.45 13.45⫾0.53 10.00⫾0.40 0.94⫾0.03 2,696.03⫾80.19 2,282.22⫾90.89 79.91⫾2.38a
a
Pairwise t tests used to calculated differences between Head Start and each of the other preschool groups. RAE⫽retinol activity equivalent. *P⬍0.05, for pairwise comparison with Head Start. **P⬍0.01, for pairwise comparison with Head Start. b
Statistical Methods Data were analyzed using the survey procedures in SAS version 9.1 (SAS Institute Inc, Cary, NC) to account for the complex survey design (18). Mean energy, macronutrient, and selected micronutrient intakes were calculated for each preschool group in addition to the percentage of children not meeting the Dietary Reference Intakes for each nutrient. Pairwise comparisons between HS and the other groups were conducted using the SURVEYREG and SURVEYFREQ procedures. Multivariate logistic regression analyses were performed to examine the odds of inadequate nutrient intake for each of the other preschool groups compared with HS. Covariates included age, race/ethnicity, poverty level, teenage pregnancy, household smoking, household size, household receipt of WIC benefits, household reference person foreign born and education level, supplement use, and dietary recall day. This
1024
July 2011 Volume 111 Number 7
study was deemed exempt under federal regulation 45 CFR §46.101(b). RESULTS Description of the Sample Compared with the other groups, HS children were more likely to be non-Hispanic black, to have a mother who was pregnant as a teenager, to live in a household receiving Special Supplemental Nutrition Program for Women, Infants, and Children benefits, and have a foreign-born household reference person. In addition, they were the least likely to be taking dietary supplements (Table 1). Children who were attending a non-HS preschool were more likely to be living in a household in which the household reference person had some college education; conversely, they were less likely to be in households below the federal poverty line and with a smoker present.
Table 3. Percentage of low-income preschool-aged children in the 1999-2004 National Health and Nutrition Examination Survey not meeting dietary guidelines for key nutrients on recall day Preschool Statusb Does not meet guideline for:a
Overall
Head Start (unweighted nⴝ184)
Non–Head Start preschool (unweighted nⴝ189)
Past preschool (unweighted nⴝ193)
No preschool (unweighted nⴝ384)
4™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™ weighted %⫾standard error ™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™3 Macronutrient Protein Total fatc Saturated fatd Micronutrient Vitamin A (RAEe) Vitamin B-1 Vitamin B-2 Vitamin B-3 Vitamin B-6 Vitamin B-9 Vitamin B-12 Vitamin C Vitamin E Calcium Magnesium Iron Zinc Copper Sodiumf Potassium Selenium
2.3⫾0.7 29.3⫾2.1 69.1⫾2.4
4.5⫾2.7 25.2⫾4.3 71.5⫾3.8
0.1⫾0.4*** 30.4⫾4.5 64.8⫾4.8
0.2⫾1.1 26.1⫾3.9 69.3⫾4.2
0.5⫾3.3 32.1⫾4.4 70.3⫾4.1
39.7⫾2.6 5.4⫾1.0 3.7⫾1.0 8.1⫾1.3 8.1⫾1.5 20.3⫾1.8 6.5⫾1.0 12.3⫾2.3 81.0⫾2.1 40.2⫾2.6 15.3⫾2.2 28.8⫾2.8 12.6⫾2.0 7.0⫾1.2 94.8⫾1.0 87.6⫾1.4 3.5⫾0.8
43.3⫾4.1 9.5⫾2.8 8.2⫾2.6 14.6⫾3.7 5.3⫾1.9 20.5⫾4.2 9.4⫾3.5 12.9⫾3.5 79.7⫾2.8 50.4⫾4.3 16.3⫾3.1 26.0⫾4.0 16.5⫾3.5 7.9⫾2.4 96.1⫾1.7 90.8⫾3.2 7.1⫾2.6
45.9⫾5.2 5.0⫾1.9 2.7⫾1.6 4.9⫾2.1*** 7.6⫾2.8 19.0⫾3.5 7.2⫾2.6 14.2⫾4.2 80.0⫾3.9 46.7⫾5.1 17.8⫾4.5 26.8⫾5.4 17.2⫾4.0 8.7⫾3.0 94.9⫾1.9 85.4⫾3.6 1.1⫾0.5***
39.3⫾4.5 2.0⫾1.0*** 2.5⫾1.2* 6.5⫾1.4*** 8.4⫾3.2 28.3⫾4.8 5.6⫾1.7 15.0⫾4.7 89.9⫾2.6*** 42.5⫾4.6 14.1⫾4.0 31.8⫾5.5 10.9⫾2.5 5.8⫾2.3 94.5⫾2.1 94.8⫾2.3 2.3⫾1.0*
35.3⫾4.3 6.1⫾1.8 3.3⫾1.5 8.2⫾2.0 9.3⫾2.2 16.1⫾2.4 5.6⫾1.5 9.3⫾2.3 76.9⫾3.7 31.4⫾3.4*** 14.2⫾3.1 29.3⫾3.5 9.7⫾2.4 6.5⫾1.8 94.7⫾1.6 83.3⫾2.6 4.1⫾1.4
a
Dietary guidelines refer to dietary reference intakes defined as recommended daily allowances or adequate intakes unless otherwise noted. Rao-Scott 2 test used to calculate differences between Head Start and each of the other preschool groups. Guideline for total fat intake defined as percent energy from fat is ⬍40% of total energy intake for 3-year-old children and less than 35% of total energy for 4- to 5-year-old children. d Guideline for saturated fat intake defined as percent energy from saturated fat is ⬍10% of total energy intake for all children. e RAE⫽retinol activity equivalent. f Guideline for sodium intake defined as below the daily tolerable Upper Level for age. *P⬍0.05, for pairwise comparison with Head Start. ***P⬍0.001, for pairwise comparison with Head Start. b c
The age distributions of HS and non-HS preschool children were comparable, whereas past-preschool children tended to be older and no-preschool children younger. The prevalence of overweight did not vary among preschool groups. Mean Dietary Intake Values All four groups reported adequate mean intakes of all key macronutrients and micronutrients except vitamin E, for which all four groups had mean intakes below the agespecific RDA (Table 2). HS children tended to report lower mean intakes of protein, saturated fat, thiamin, riboflavin, niacin, calcium, phosphorous, iron, and selenium than children in the other three categories. The greatest differences in intake were observed between HS and no-preschool children; no-preschool children had significantly higher mean intakes of protein, fat, riboflavin, calcium, and phosphorous. There were no significant differences at the mean level between HS children and non-HS preschool children. No differences were observed at the mean energy level, suggesting that individual
groups were not systematically under- or overreporting intakes. Percentage of Children Not Meeting Recommended Intakes Overall, the nutrients for which the greatest percentage of children did not meet the RDA included folate (20.5%), vitamin A (39.7%), vitamin E (79.7%), calcium (40.2%), iron (28.8%), and potassium (90.8%) (Table 3). In comparison to the other three groups, HS had more children not meeting the requirements for protein, riboflavin, calcium. The prevalence of low niacin intake was 14.6% among HS children, compared to 4.9% in non-HS and 6.5% in pastpreschool children. Fifty percent of HS children did not meet the daily calcium requirements vs only 31.4% of no-preschool children. Many of these differences were statistically significant in pairwise comparisons between HS and other preschool groups (Table 3). Mutivariable Analyses Multivariate analyses showed that children in HS were between 3.1 and 9.6 times more likely to not meet vita-
July 2011 ● Journal of the AMERICAN DIETETIC ASSOCIATION
1025
Table 4. Results of unadjusted and adjusted logistic regression analyses for association between preschool status and not meeting dietary guidelines in the 1999-2004 National Health and Nutrition Examination Surveyab Preschool Status
Intake Macronutrient Protein Unadjusted ORd (95% CIe) Adjusted OR (95% CI) Total fat Unadjusted OR (95% CI) Adjusted OR (95% CI) Saturated fat Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin A (RAEf) Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin B-1 Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin B-2 Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin B-3 Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin B-6 Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin B-9 Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin B-12 Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin C Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin E Unadjusted OR (95% CI) Adjusted OR (95% CI) Calcium Unadjusted OR (95% CI) Adjusetd OR (95% CI) Magnesium Unadjusted OR (95% CI) Adjusted OR (95% CI) Iron Unadjusted OR (95% CI) Adjusted OR (95% CI) Zinc Unadjusted OR (95% CI) Adjusted OR (95% CI) Copper Unadjusted OR (95% CI) Adjusted OR (95% CI)
1026
Head Startc (unweighted nⴝ184)
Non–Head Start preschool (unweighted nⴝ189)
Past preschool (unweighted nⴝ193)
No preschool (unweighted nⴝ384)
1.00 1.00
0.08 (0.01, 0.53) 0.07 (0.01, 0.69)
0.24 (0.04, 1.54) 0.20 (0.02, 1.76)
0.72 (0.17, 3.10) 0.64 (0.11, 3.55)
1.00 1.00
1.30 (0.78, 2.16) 1.54 (0.87, 2.74)
1.05 (0.53, 2.06) 1.19 (0.55, 2.54)
1.40 (0.74, 2.67) 2.38 (1.17, 4.85)
1.00 1.00
0.73 (0.46, 1.17) 0.77 (0.46, 1.29)
0.90 (0.50, 1.62) 0.84 (0.45, 1.60)
0.94 (0.54, 1.64) 0.89 (0.50, 1.56)
1.00 1.00
1.11 (0.71, 1.73) 1.24 (0.73, 2.10)
0.85 (0.51, 1.40) 0.89 (0.49, 1.64)
0.71 (0.42, 1.20) 0.89 (0.51, 1.54)
1.00 1.00
0.50 (0.17, 1.47) 0.32 (0.10, 1.00)
0.20 (0.07, 0.53) 0.17 (0.05, 0.57)
0.62 (0.24, 1.63) 0.55 (0.15, 2.07)
1.00 1.00
0.30 (0.08, 1.11) 0.10 (0.03, 0.38)
0.29 (0.08, 1.12) 0.21 (0.05, 0.88)
0.37 (0.12, 1.16) 0.35 (0.11, 1.15)
1.00 1.00
0.30 (0.14, 0.67) 0.24 (0.11, 0.54)
0.41 (0.23, 0.73) 0.31 (0.13, 0.78)
0.53 (0.23, 1.21) 0.36 (0.11, 1.11)
1.00 1.00
1.46 (0.52, 4.13) 1.26 (0.45, 3.57)
1.63 (0.50, 5.30) 1.25 (0.35, 4.55)
1.81 (0.74, 4.46) 1.72 (0.56, 5.23)
1.00 1.00
0.91 (0.43, 1.92) 0.68 (0.32, 1.45)
1.54 (0.75, 3.17) 1.27 (0.61, 2.63)
0.75 (0.40, 1.42) 0.73 (0.36, 1.50)
1.00 1.00
0.74 (0.22, 2.54) 0.61 (0.14, 2.74)
0.57 (0.20, 1.64) 0.57 (0.18, 1.84)
0.57 (0.21, 1.57) 0.68 (0.22, 2.09)
1.00 1.00
1.11 (0.49, 2.53) 1.05 (0.44, 2.50)
1.18 (0.60, 2.33) 0.90 (0.43, 1.91)
0.69 (0.30, 1.57) 0.64 (0.26, 1.54)
1.00 1.00
1.02 (0.56, 1.84) 0.79 (0.46, 1.36)
2.27 (1.34, 3.85) 1.83 (1.02, 3.28)
0.85 (0.53, 1.35) 0.73 (0.42, 1.28)
1.00 1.00
0.86 (0.51, 1.46) 0.92 (0.49, 1.73)
0.73 (0.47, 1.13) 0.79 (0.47, 1.33)
0.45 (0.28, 0.73) 0.55 (0.31, 0.96)
1.00 1.00
1.11 (0.50, 2.50) 1.08 (0.49, 2.39)
0.85 (0.37, 1.91) 0.76 (0.36, 1.58)
0.85 (0.43, 1.71) 1.04 (0.49, 2.22)
1.00 1.00
1.04 (0.56, 1.93) 0.79 (0.39, 1.60)
1.33 (0.73, 2.40) 1.16 (0.59, 2.29)
1.18 (0.67, 2.09) 1.02 (0.50, 2.09)
1.00 1.00
1.05 (0.54, 2.05) 0.77 (0.31, 1.93)
0.62 (0.35, 1.11) 0.48 (0.21, 1.11)
0.54 (0.25, 1.18) 0.51 (0.19, 1.41)
1.00 1.00
1.12 (0.36, 3.52) 0.95 (0.32, 2.84)
0.72 (0.28, 1.86) 0.70 (0.24, 2.02)
0.81 (0.33, 2.03) 0.85 (0.32, 2.32) (continued)
July 2011 Volume 111 Number 7
Table 4. Results of unadjusted and adjusted logistic regression analyses for association between preschool status and not meeting dietary guidelines in the 1999-2004 National Health and Nutrition Examination Surveyab (Continued) Intake
Sodium Unadjusted OR (95% CI) Adjusted OR (95% CI) Potassium Unadjusted OR (95% CI) Adjusted OR (95% CI) Selenium Unadjusted OR (95% CI) Adjusted OR (95% CI)
Preschool Status Head Startc (unweighted nⴝ184)
Non–Head Start preschool (unweighted nⴝ189)
Past preschool (unweighted nⴝ193)
No preschool (unweighted nⴝ384)
1.00 1.00
0.70 (0.36, 1.36) 0.75 (0.37, 1.54)
1.62 (0.84, 3.13) 1.59 (0.75, 3.35)
1.30 (0.74, 2.29) 1.25 (0.67, 2.33)
1.00 1.00
0.59 (0.21, 1.65) 0.54 (0.17, 1.72)
1.85 (0.53, 6.48) 1.93 (0.55, 6.81)
0.51 (0.23, 1.12) 0.72 (0.28, 1.89)
1.00 1.00
0.15 (0.04, 0.54) 0.15 (0.03, 0.74)
0.31 (0.12, 0.82) 0.33 (0.09, 1.13)
0.56 (0.20, 1.57) 0.50 (0.16, 1.58)
a
Dietary guidelines refer to Dietary Reference Intakes. Adjusted for age; race/ethnicity; poverty level; teenage pregnancy; household smoking; household size; household receipt of Special Supplemental Nutrition Program for Women, Infants, and Children benefits; household reference person foreign born and education level; supplement use; and dietary recall day. c Reference category. d OR⫽odds ratio. e CI⫽confidence interval. f RAE⫽retinol activity equivalent. b
mins B-1 to B-3 requirements as compared to non-HS and past-preschoolers (Table 4). Similarly, they were 1.83 times more likely to not meet calcium requirements than no-preschoolers (P⬍0.05). HS children tended to have a higher risk of inadequate intake of protein, saturated fat, thiamin, riboflavin, niacin, vitamin B-12, calcium, magnesium, zinc, copper, and selenium, although in many cases these differences were not significant. Mean Serum Micronutrient Values To help assess the validity of the observed differences in reported dietary intake among the preschool groups, mean levels of all serum micronutrients for which there were available NHANES data were examined (Table 5). Consistent with parental reports, HS children had the lowest mean serum levels for approximately one half of the vitamins and minerals, including folate, vitamin E, selenium, vitamin A, retinol stearate, and gamma tocopherol. Although intake data from one 24-hour recall is not sufficient for direct comparison to serum levels, a poorer serum profile would seem to suggest habitually low intakes and a lower-quality diet on the whole. Thus, the observation that HS children frequently had lower mean serum vitamin levels supports the validity of our findings on reported dietary intake and suggests that observed differences did not result from differential recall and misclassification. DISCUSSION These data demonstrate that, on average, children in HS consume adequate amounts of all key nutrients except vitamin E. However, compared to other low-income children of the same age, HS children appear to be at greater risk for not meeting the RDAs for several vitamins and
minerals, including thiamin, riboflavin, niacin, selenium, and calcium. The high percentage of HS children not meeting calcium requirements (40.2%) is of particular concern given the relationship of dairy consumption to height (19), weight gain (20), and bone health (21,22). Findings in the parent-reported intake data appear to be substantiated by observations at the serum level, which show that HS participants have generally poorer serum vitamin profiles. No differences in the prevalence of overweight among the preschool groups were observed, although children in our low-income sample had a higher percentage of overweight and at risk of overweight than the general population of 2- to 5-year-olds. Previous research on HS nutrition has focused on the quality of school meals only or has used children of nonsimilar socioeconomic backgrounds for comparison. The results of this study add to the literature by assessing intake throughout the day and by comparing micronutrient intake of HS children to that of other children living in poverty. For example, Oakley and colleagues (23) analyzed menus at Child and Adult Care Food Program (CACFP)-funded centers. They found that whereas nearly all menus met the meal-pattern requirements established by the HS Performance Standards (12), following these guidelines did not guarantee consistent nutrition quality of planned meals. In fact, menus from CACFP centers failed to meet one third of the RDAs for energy, vitamins B-12 and E, calcium, iron, and zinc (23). Other studies have also found that average nutrient intakes at school for children in HS or CACFP-funded programs are not consistent with HS standards. In particular, meals served in HS classrooms have been found to be substandard for energy, saturated fat, calcium, vitamin E, sodium, iron, and zinc requirements (24,25). In general, menus and dietary intakes in childcare centers have
July 2011 ● Journal of the AMERICAN DIETETIC ASSOCIATION
1027
Table 5. Adjusted mean⫾standard error (SE) serum micronutrient levels by preschool status in the 1999-2004 National Health and Nutrition Examination Survey Preschool Statusa
Macronutrient Vitamin B-6 (mg) 1 Cohort (n⫽201) Vitamin B-9 (folate) (ng/mL) 3 Cohorts (n⫽736) Folate, RBCb (ng/mL) 3 Cohorts (n⫽762) Vitamin B-12 (g) 3 Cohorts (n⫽737) Vitamin E (g/dL) 2 Cohorts (n⫽448) Iron (g/dL) 3 Cohorts (n⫽749) Selenium (ng/mL) 1 Cohort (n⫽205) Vitamin A (g/dL) 2 Cohorts (n⫽450) Ret stearate (g/dL) 2 Cohorts (n⫽347) Ret palmitate (g/dL) 2 Cohorts (n⫽379) ␥-Tocopherol (g/dL) 2 Cohorts (n⫽413) ␣-Carotene (g/dL) 1 Cohort (n⫽239) Trans beta carotene (g/dL) 1 Cohort (n⫽239) Cis beta carotene (g/dL) 1 Cohort (n⫽239) Vitamin D (ng/mL) 2 Cohorts (n⫽282)
Non–Head Start preschool (unweighted nⴝ189)
Overall
Head Start (unweighted nⴝ184)
Past preschool (unweighted nⴝ193)
No preschool (unweighted nⴝ384)
4™mean⫾SE™3
4™™™™™™™™™™™™™™™™™™™™™™ weighted mean⫾SE ™™™™™™™™™™™™™™™™™™™™™™™ 3
44.77⫾4.20
46.68⫾4.16
50.97⫾9.83
49.95⫾8.35
38.01⫾3.95
17.99⫾0.52
18.35⫾0.70
18.07⫾0.86
18.71⫾0.96
17.36⫾0.84
283.76⫾5.83
268.84⫾6.24
297.26⫾15.22
280.39⫾12.61
284.64⫾5.99*
815.87⫾21.12
874.76⫾48.83
789.40⫾32.59
816.24⫾33.48
805.38⫾33.27
799.01⫾16.00
730.98⫾20.14
854.14⫾25.75**
809.31⫾44.85
790.29⫾20.24*
74.36⫾1.38
72.77⫾4.37
77.37⫾3.07
72.51⫾2.97
74.42⫾2.32
106.99⫾1.27
101.25⫾2.88
108.05⫾2.15*
113.04⫾2.89*
106.37⫾1.46
33.61⫾0.50
31.86⫾1.17
35.19⫾0.83*
33.37⫾0.97
33.51⫾0.58
0.56⫾0.07
0.44⫾0.04
0.68⫾0.23
0.46⫾0.04
0.60⫾0.07*
2.28⫾0.18
2.12⫾0.18
2.78⫾0.56
1.95⫾0.17
2.24⫾0.21
205.35⫾14.20
233.64⫾10.97
207.49⫾28.67
2.84⫾0.31
3.52⫾0.81
2.95⫾0.55
3.50⫾0.75
2.26⫾0.16
14.09⫾0.69
14.78⫾1.37
16.30⫾1.36
14.49⫾2.22
12.68⫾0.89
0.70⫾0.03
0.73⫾0.06
0.80⫾0.06
0.71⫾0.09
0.65⫾0.04
28.76⫾0.86
28.34⫾1.67
27.26⫾1.07
29.90⫾1.19
28.74⫾1.19
216.67⫾8.25
216.53⫾7.89
a
Pairwise t tests used to calculated differences between Head Start and each of the other preschool groups. RBC⫽red blood cell. *P⬍0.05, for pairwise comparison with Head Start. **P⬍0.01, for pairwise comparison with Head Start. b
been shown to be lacking in energy, iron, niacin, thiamin, calcium, vitamin B-6, vitamin A, folic acid, and protein (26-28). Worobey and colleagues (29) found that low-income children attending an HS preschool consumed similar levels of protein, carbohydrate, and fat as middleincome children at a private preschool in the same area. However, they also found that the home diets of the HS children were higher in fat and carbohydrates. Thus, studies to date have reported mixed results on the quality of meals served in Head Start and preschool programs. This discrepancy is likely attributable to differences in the data sources, dietary measurement techniques, and comparison groups. This study is the first to compare the diets of HS children to those of other low-income preschool-aged children and to show that children in HS are at higher risk for not
1028
July 2011 Volume 111 Number 7
meeting the RDAs for several vitamins and minerals. Although it could not be determined whether these findings are the consequence of suboptimal nutrition at school or at home, the results of this study present opportunities for intervention. If the observed differences in diet quality at the preschool level are a result of the food served at HS programs, then stronger HS policies, increased CACFP funding, and additional staff training may help to improve the nutritional intake of children in HS. Although HS centers consult dietetics practitioners more frequently than non-HS centers when designing menus (30), these dietetics practitioners may not have the capacity to offer healthier choices without stronger policies or expanded budgets. If, on the other hand, the differences are a result of food served at home, then greater communication between HS centers and parents
in the form of nutrition awareness and education may improve nutritional intake (31,32). The observation that HS children were more likely to have foreign-born heads of household also warrants discussion. Households with foreign-born heads may have certain cultural or religious practices that influence diet at home, and foreign-born parents may have different beliefs or expectations from government programs like HS. In a survey of obesity prevention in HS, 58% of HS directors cited cultural beliefs about food as a barrier to healthy eating by HS children (33). Although diets in these households may be healthy and meet daily requirements, preschool educators may find it challenging to educate these children and families about nutrition practices and quality. Given this observation, nutrition information relayed to children and parents should be mindful of differences in cultural practices and foods while emphasizing the importance of adhering to a healthy diet. This study illustrates important differences in diet quality by preschool status, but study limitations must be acknowledged. First, a single 24-hour recall was used to evaluate preschoolers’ dietary intakes. Although a single recall has been shown to be sufficient for calculating mean food and nutrient intakes on a population level, the use of multiple recalls is generally preferred for estimating the prevalence of high or low intakes (34). The absence of multiple recall days for all members of the NHANES sample prevented us from being able to estimate usual intake. However, this limitation would not have been expected to cause differential biases between preschool groups because the same methods of calculating inadequate intake were used for all groups. Second, the use of parents as proxies for the measurement of preschoolers’ dietary intake may increase the likelihood of misreporting. Parents with children in preschool or child care may be less informed about the types and quantities of food their child consumes outside of the home. Whereas the possibility of differential misreporting exists, all preschool groups had similar mean caloric intakes with relatively narrow standard errors, suggesting that no group had systematically under- or overreported children’s intakes. Third, variability in the nutrition environment and diet quality within the category of non-HS preschools may be sizeable, such that treating these schools as a unit may not be appropriate. In fact, this limitation may explain why the fewest number of differences were observed between this group and HS children. In addition, children classified as past preschoolers may have attended either Head Start or non-HS programs. Fourth, students who attended HS part day vs full day could not be differentiated nor could foods consumed at home vs school. Fifth, observed dietary intakes could not be directly related to serum levels because data on most serum vitamins were not available and because usual intakes could not be estimated. Sixth, a greater percentage of HS students lived in households with incomes below the federal poverty level, which may influence access to healthy foods at home. Finally, the problem of multiple comparisons may have been present in the results. To help distinguish highly significant findings from weaker ones, separate indicators for varying levels of significance are indicated.
CONCLUSIONS Children living in poverty are at high risk for nutritional deficiencies, obesity, and subsequent inequities in health (ie, hypertension, metabolic syndrome, and diabetes) (6,8). Despite federal efforts to improve nutrition in lowincome children through the HS program, the diets of HS preschoolers do not appear to be on par with those of other preschool-aged children; in general, HS children appear to have poorer quality diets than non-HS children of the same socioeconomic status. Because nutrition provides an important foundation for academic performance and overall health, ensuring that HS children receive a diet that is as equally nutritious as that of other children is imperative for success. Further research is needed to delineate the source of the observed differences in diet quality and ultimately to design strategies for enhancing the diets of HS children both at home and in school. STATEMENT OF POTENTIAL CONFLICT OF INTEREST: No potential conflict of interest was reported by the authors. FUNDING/SUPPORT: This project was independently performed. The authors have no funding sources to disclose. References 1. Polhamus B, Dakenius K, Thompson D, Scanlon K, Borland E, Smith B, Grummer-Strawn L. Pediatric Nutrition Surveillance 2003 Report. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2004. 2. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States, 19992004. JAMA. 2006;295:1549-1555. 3. Alaimo K, Olson CM, Frongillo EA, Briefel RR. Food insufficiency, family income, and health in US preschool and school-aged children. Am J Public Health. 2001;91:781-786. 4. Kotch J, Shackelford J. The nutritional status of low-income preschool children in the United States: A review of the literature. 1989. Food Research and Action Center Web site. http://frac.org. Accessed April 4, 2009. 5. Rose D, Smallwood D, Blaylock J. Socioeconomic factors associated with the iron intake of preschoolers in the United States. Nutr Res. 1995;15:1298-1309. 6. Birch LL. Development of food preferences. Annu Rev Nutr. 1999;19: 41-62. 7. Birch LL. Effects of peer models’ food choices and eating behaviors on preschoolers’ food preferences. Child Dev. 1980;51:489-496. 8. Skinner JD, Carruth BR, Bunds W, Ziegler PJ. Children’s food preferences: A longitudinal analysis. J Am Diet Assoc. 2002;102:16381647. 9. Hoerr S, Utech AE, Ruth E. Child control of food choices in Head Start families. J Nutr Educ Behav. 2005;37:185-190. 10. Capizzano J, Adams G. The Hours That Children Under Five Spend in Child Care: Variation Across States. Washington, DC: Urban Institute; 2000. 11. Wright JD, Borrud LG, McDowell MA, Wang C, Radimer K, Johnson CL. Nutrition assessment in the National Health and Nutrition Examination Survey 1999-2002. J Am Diet Assoc. 2007;107:822-829. 12. Head Start Program performance standards and other regulations: Part 1304.23: Child nutrition. Office of Head Start Web site. http:// eclkc.ohs.acf.hhs.gov/hslc. Accessed April 10, 2009. 13. National Health and Nutrition Examination Survey. National Center for Health Statistics, Centers for Disease Control and Prevention Web site. www.cdc.gov/nchs/nhanes.htm. Accessed April 7, 2009. 14. USDA Food and Nutrient Database for Dietary Studies. US Department of Agriculture Web site. http://www.usda.gov/wps/portal/usda/ usdahome. Accessed April 7, 2009. 15. Kleinman RE. Pediatric Nutrition Handbook. 6th ed. Elk Grove Village, IL: American Academy of Pediatrics; 2009:1294-1295. 16. Dietary Guidelines for Americans 2005. US Department of Agricul-
July 2011 ● Journal of the AMERICAN DIETETIC ASSOCIATION
1029
17. 18.
19. 20. 21. 22. 23. 24. 25.
ture Web site. http://www.health.gov/dietaryguidelines. Accessed July 6, 2009. A SAS Program for the CDC Growth Charts. Atlanta, GA: Centers for Disease Control and Prevention; 2005. Gossett JM, Jo CH, Simpson PM. US Health and Nutrition: SAS survey procedures and NHANES. Paper presented at 31st Annual SAS Users Group International Conference, March 26-29, 2006; San Francisco, CA. http://www2.sas.com/proceedings/sugi31/140-31.pdf. Accessed April 7, 2009. Wiley AS. Does milk make children grow? Relationships between milk consumption and height in NHANES 1999-2002, Am J Hum Biol. 2005;17:425-441. Moore LL, Bradlee ML, Gao D, Singer MR. Low dairy intake in early childhood predicts excess body fat gain. Obes Res. 2006;14:1010-1018. Black RE, Williams SM, Jones IE, Goulding A. Children who avoid drinking cow’s milk have lower dietary calcium intakes and poor bone health. Am J Clin Nutr. 2002;76:675-680. Teegarden D, Lyle RM, Proulx WR, Johnston CC. Previous milk consumption is associated with greater bone density in young women. Am J Clin Nutr. 1999;69:1014-1017. Oakley CB, Bomba AK, Knight KB, Byrd SH. Evaluation of menus planned in Mississippi child-care centers participating in the Child and Adult Care Food Program. J Am Diet Assoc. 1995;95:765-768. Fox MK, Glantz FB, Geitz L, Burstein N. Early Childhood and Child Care Study: Nutritional Assessment of the CACFP. Cambridge, MA: Abt Associates Inc; 1997. Bollella MC, Spark A, Boccia LA, Nicklas TA, Pittman BP, Williams
1030
July 2011 Volume 111 Number 7
26. 27. 28. 29. 30.
31. 32.
33. 34.
CL. Nutrient intake of Head Start children: Home vs school. J Am Coll Nutr. 1999;18:108-114. Briley M, Buller A, Roberts-Gray CC, Sparkman A. What is on the menu at the childcare center? J Am Diet Assoc. 1989;89:771-774. Briley M, Roberts-Gray C, Rowe S. What can children learn from the menu at the child-care center? J Community Health. 1993;18:363-377. Padget A, Briley ME. Dietary intakes at child-care centers in central Texas fail to meet Food Guide Pyramid recommendations. J Am Diet Assoc. 2005;105:790-793. Worobey J, Worobey HS, Adler AL. Diet, activity, and BMI in preschool-aged children: Differences across settings. Ecol Food Nutr. 2005;44:455-466. Gupta RS, Pascoe JM, Blanchard TC, Langkamp D, Duncan PM, Gorski PA, Southward LH. Child health in child care: A multi-state survey of Head Start and non-Head Start child care directors. J Pediatr Health Care. 2009;23:143-149. Briley ME, Jastrow S, Vickers J, Roberts-Gray C. Dietary intake at child-care centers and away: Are parents and care providers working as partners or at cross-purposes? J Am Diet Assoc. 1999;99:950-954. Spurrier NJ, Magarey AA, Golley R, Curnow F, Sawyer MG. Relationships between the home environment and physical activity and dietary patterns of preschool children: A cross-sectional study. Int J Behav Nutr Phys Act. 2008;5:31-42. Hughes CC, Gooze RA, Finkelstein DM, Whitaker RC. Barriers to obesity prevention in Head Start. Health Affairs. 2010;29:454-462. Willett W. Nutritional Epidemiology, 2nd ed. New York, NY: Oxford University Press; 1998:54.
Dietary Intake in Head Start vs Non–Head Start Preschool-Aged Children: Results from the 1999-2004 National Health and Nutrition Examination Survey EMILY M. BUCHOLZ, MPH; MAYUR M. DESAI, PhD, MPH; MARJORIE S. ROSENTHAL, MD, MPH
ABSTRACT Objective To determine whether dietary intakes of children enrolled in Head Start programs differ from those of children not attending preschool or children in non–Head Start programs. Design Using data from the 1999-2004 National Health and Nutrition Examination Survey, low-income, 3- to 5-year-old children were categorized into one of four preschool groups: Head Start (n⫽184), non–Head Start (n⫽189), past preschool (n⫽193), and no preschool (n⫽384). Total nutrient intakes were calculated using 24-hour parental recalls. Statistical analyses performed Mean macronutrient and micronutrient intakes were compared across groups and the percentage of children not meeting Recommended Dietary Allowances (RDAs) were calculated. Multivariate logistic regression was used to evaluate the relationship between preschool group and likelihood of not meeting dietary guidelines. Results Many children did not meet the RDA for folate (20.5%), vitamin A (39.7%), vitamin E (79.7%), calcium (40.2%), iron (28.8%), and potassium (90.8%). Compared with the other preschool groups, Head Start children had lower mean protein, saturated fat, riboflavin, calcium, and phosphorous intakes. The greatest differences in intake were observed between Head Start participants and no-preschool children. Multivariate analyses demonstrated an association between Head Start and inadequate intake of protein, thiamin, riboflavin, niacin, calcium, and selenium.
E. M. Bucholz is a medical student, M. M. Desai is an assistant professor, Division of Chronic Disease Epidemiology, and M. S. Rosenthal is an associate research scientist, Department of Pediatrics, and assistant director, Robert Wood Johnson Foundation Clinical Scholars Program, Yale School of Medicine, New Haven, CT. Address correspondence to: Emily M. Bucholz, MPH, Yale School of Medicine, 333 Cedar St, New Haven, CT 06511. E-mail: [email protected] Manuscript accepted: January 24, 2011. Copyright © 2011 by the American Dietetic Association. 0002-8223/$36.00 doi: 10.1016/j.jada.2011.04.009
© 2011 by the American Dietetic Association
Conclusions Compared with other low-income children, those in Head Start programs appear to be at greater risk for not meeting the RDA for several key vitamins and minerals. These differences in diet quality may present an opportunity for Head Start programs to enhance nutrition in this student population. J Am Diet Assoc. 2011;111:1021-1030.
P
reschool-aged children are at risk for inadequate nutrition and, among low-income preschoolers, the risk is even greater: 30% of preschool-aged children living below the poverty line are overweight or at risk of overweight (1) compared with 26% of all 2- to 5-year-olds (2). Children living in poverty are also at higher risk of micronutrient deficiencies, particularly zinc and iron (35). Given that the preschool years are when children begin exercising greater control over what they eat and establish eating patterns that may last a lifetime (6-9), the environmental context of a preschool-aged child may play a key role in lifetime nutrition. Nationwide, 60% of all US children receive at least some child care outside of the home, and one half of these children spend 35 hours or more each week in out-ofhome child care (10). As a result, many children consume more than one half of their daily food intake in out-of-home care. Head Start (HS), a federally funded preschool program serving low-income children, aims to “promote school readiness by enhancing the social and cognitive development of children through the provision of educational, health, nutritional, social and other services to enrolled children and families” (11). This program may serve as an opportunity to enhance nutritional intake. All HS programs are required to comply with federal Program Performance Standards that address the identification of children’s nutrition needs, menu planning and dietary requirements for meals served in the classroom, feeding practices and nutrition education, food safety and sanitation, and family education around child nutrition (12). Despite risk of inadequate nutrition for low-income preschool-aged children and the emphasis HS places on health, there is a paucity of literature on the nutritional benefits of HS participation. The purpose of this study was to examine, using a nationally representative sample of low-income children, whether dietary intakes of children enrolled in HS programs differ from those of chil-
Journal of the AMERICAN DIETETIC ASSOCIATION
1021
dren not attending preschool or children in non-HS preschool programs. METHODS Data Source Data from the National Health and Nutrition Examination Survey (NHANES) were used to assess dietary intake in US preschool-aged children. Conducted by the National Center for Health Statistics, NHANES is a nationally representative, cross-sectional survey of the civilian, noninstitutionalized population of the United States. It uses stratified, multistage probability sampling to oversample low-income individuals, certain age groups, and races/ethnicities (13). For this study, data from three NHANES cohorts was combined: 1999-2000, 2001-2002, and 2003-2004. NHANES collects detailed demographic, socioeconomic, dietary, and health-related data through in-home interviews and clinical assessments at a mobile examination center. For children younger than age 6 years, these interviews were conducted using parent or guardian proxies. A detailed description of these methods is available elsewhere (11,13). Selection of Participants and Measures The sample consisted of all children aged 3 to 5 years living in poverty (200% of the federal poverty line and below) who had available dietary intake data (N⫽950). We limited the sample to children whose annual family income fell at or below 200% of the federal poverty line for household size and year to compare children of similar socioeconomic status while still including the majority of children in HS programs. Preschool Status Participants were categorized into one of four groups according to history of preschool attendance using the following questions: “Is sample person [SP] now attending HS?” “Does SP now attend day care or preschool?” and “Did SP ever attend day care or preschool?” The four categories used were HS (children currently in HS), non-HS preschool (children in a non-HS preschool/child care program), past preschool (children who previously attended but no longer attend a preschool/child care program), and no preschool (children who have never attended a preschool/child care program). Dietary Intake and Serum Micronutrient Levels The NHANES total nutrient databases were used to estimate nutrient intakes for each group of children. Information on individual foods consumed by participants was collected by trained NHANES personnel using 24-hour parental recalls and then converted to total nutrient intakes using the US Department of Agriculture’s Food and Nutrient Database for Dietary Studies (14). Before 2003, a single dietary interview was administered during the mobile examination center exam to assess individual food and total nutrient consumption on 1 day only. Since then, NHANES has added an additional 24-hour recall to collect data on 2 days of intake, using
1022
July 2011 Volume 111 Number 7
both in-person and telephone interviews. Due to differences in the number of days for which dietary information was available for children in this study, analyses were restricted to the 1-day in-person data collected in the mobile examination center. To estimate the prevalence of children not meeting the recommended intake levels for specific macro- and micronutrients, total nutrient intakes were compared to age-specific standards. The standards used were Dietary Reference Intakes for protein and all reported micronutrients as defined by the Institute of Medicine’s Food and Nutrition Board guidelines (15). Because lowincome children are already at high risk of nutrient deficiencies, the more stringent criteria Recommended Daily Allowances (RDAs) were employed, as opposed to Estimated Average Requirements. RDAs are calculated to meet the need of 97% to 98% of individuals whereas Estimated Average Requirements only meet the needs of 50% of individuals. When RDAs were not available, Adequate Intakes were used. In the case of sodium, the tolerable Upper Level was used as a cut-off because of concerns over excess sodium. Guidelines for fat intakes were defined as not exceeding 35% to 40% of total energy from fat and 10% from saturated fat based on the US Department of Agriculture’s Dietary Guidelines for Americans 2005 (16). When available, serum concentrations of micronutrients were examined to determine whether differences in reported dietary intakes between preschool groups could be explained by differential misreporting. Methods describing the measurement of these variables and their associated errors are published elsewhere (13). Because NHANES collected different serum micronutrients in each of the three cohorts, the sample size and number of cohorts used for the analysis of each vitamin or mineral are indicated. Covariates The association between preschool group and several demographic, socioeconomic, and health-related variables was examined. These variables included child sex, age, race/ethnicity, insurance status, birthplace, and body mass index classification (overweight: ⬎95th percentile, at risk for overweight: ⬎85th to ⬍95th percentile, or normal: ⬍85th percentile) (17). Consistent with NHANES data collection, race/ethnicity was categorized as nonHispanic white, non-Hispanic black, Mexican American, and other. Socioeconomic variables included household size, presence of adult smoker(s) in the household, household poverty level, household Special Supplemental Nutrition Program for Women, Infants, and Children and/or food stamp receipt, the education level and birthplace of the household reference person, whether the child’s mother was a teenager while pregnant, smoked during pregnancy, or breastfed the child as an infant, and the child’s use of dietary supplements. Finally, to determine the possibility of differential recall based on day of the week, a dummy variable indicating weekend vs weekday recall was examined. Covariates associated with preschool enrollment (at the P⬍0.10 level) were included in the final multivariate models.
Table 1. Characteristics of US low-income preschool-aged children aged 3 to 5 years in the 1999-2004 National Health and Nutrition Examination Survey by preschool status (N⫽950)a Preschool Status
Variable
Overall
Head Start (unweighted nⴝ184)
Non–Head Start preschool (unweighted nⴝ189)
Past preschool (unweighted nⴝ193)
No preschool (unweighted nⴝ384)
P valueb
4™™™™™™™™™™™™™™™™™™ weighted %⫾standard error ™™™™™™™™™™™™™™™™™™™3 Child variables Girl Age (y) 3 4 5 Race/ethnicity Non-Hispanic white Non-Hispanic black Mexican American other Child born outside US Medical insurance Private Medicaid/SCHIPc Other insurance Uninsured Household variables Below federal poverty line Household size (including child) 2 3-4 5 or more Presence of adult smoker Special Supplemental Nutrition Program for Women, Infants, and Children benefits received Food Stamps received Household reference person born outside US Household reference person education level Less than high school High school Some college College graduate Pregnancy variables Teenage pregnancy Mother smoked while pregnant Child breastfed as infant Child health status variables Body mass index Normal (⬍85%) At risk of overweight Overweight (⬎95%) Supplement use Other variables Recall day of the week Weekend Weekday
51.0⫾1.9
52.7⫾4.4
46.7⫾5.4
56.0⫾4.9
49.7⫾3.5
0.568 ⬍0.001
35.8⫾2.1 32.4⫾2.3 31.8⫾1.8
29.1⫾5.5 40.3⫾4.6 30.5⫾5.2
28.2⫾3.6 42. ⫾4.7 29.7⫾5.3
25.0⫾4.1 26.4⫾4.2 48.5⫾4.5
48.7⫾3.5 27.6⫾4.5 23.7⫾3.1
48.2⫾3.9 18.6⫾2.2 19.1⫾2.5 14.2⫾2.4 3.3⫾0.7
33.6⫾6.6 37.3⫾5.7 20.3⫾4.2 8.8⫾3.4 3.6⫾1.2
52.5⫾5.6 21.2⫾2.8 11.8⫾2.5 14.6⫾4.0 2.8⫾1.2
53.4⫾7.0 20.2⫾4.3 11.6⫾2.7 14.8⫾4.3 2.5⫾1.7
48.2⫾5.7 9.2⫾1.8 26.9⫾3.8 15.6⫾3.2 4.0⫾1.1
30.9⫾2.3 47.8⫾2.5 8.2⫾1.2 13.1⫾1.8
19.8⫾3.6 57.0⫾4.2 8.7⫾2.7 14.6⫾3.3
35.3⫾5.2 47.5⫾5.4 9.1⫾3.4 8.1⫾2.5
37.3⫾5.1 41.6⫾5.0 7.0⫾2.5 14.0⫾3.9
28.9⫾3.0 48.3⫾4.2 8.1⫾1.6 14.7⫾2.8
50.9⫾2.0
61.6⫾4.4
40.0⫾4.9
45.2⫾4.7
56.1⫾3.3
5.3⫾0.9 47.3⫾2.8 47.4⫾3.0 31.9⫾3.4
5.4⫾2.1 48.9⫾5.1 45.7⫾4.8 30.3⫾5.1
8.2⫾2.8 55.4⫾4.8 36.4⫾5.1 20.8⫾4.7
6.7⫾2.0 58.8⫾5.4 34.5⫾4.8 33.1⫾5.1
2.9⫾1.2 35.6⫾3.5 61.5⫾3.8 37.9⫾5.3
0.040
46.3⫾2.7 66.5⫾3.6
55.6⫾4.9 72.6⫾5.3
38.6⫾5.4 65.0⫾7.6
41.5⫾3.8 65.6⫾6.7
49.9⫾4.4 65.4⫾5.6
0.049 0.860
24.2⫾2.6
32.3⫾3.9
17.2⫾3.9
17.7⫾3.7
22.7⫾3.6
⬍0.001 0.004
10.3⫾1.2 57.0⫾2.9 24.9⫾2.8 7.9⫾1.8
9.7⫾2.5 62.1⫾6.6 26.6⫾6.7 1.6⫾0.8
4.4⫾1.3 55.1⫾5.0 32.7⫾4.4 8.0⫾2.8
5.3⫾1.8 64.1⫾5.5 23.3⫾4.1 7.3⫾2.6
16.9⫾3.1 51.8⫾3.9 20.7⫾5.2 10.7⫾3.4
20.0⫾1.8 26.2⫾3.5 50.8⫾3.1
28.9⫾4.2 19.0⫾5.5 44.1⫾4.3
23.7⫾4.5 24.8⫾5.4 50.0⫾6.0
18.3⫾3.0 24.5⫾3.3 53.2⫾5.9
15.8⫾2.6 30.6⫾6.4 52.4⫾4.6
70.6⫾2.4 12.5⫾1.4 16.9⫾1.9 34.8⫾2.6
68.9⫾5.5 14.5⫾3.2 16.6⫾4.5 22.2⫾3.7
69.9⫾4.8 11.9⫾3.7 18.2⫾4.2 40.9⫾4.7
68.0⫾5.5 12.8⫾2.4 19.2⫾4.0 40.9⫾4.2
73.1⫾3.1 12.0⫾2.6 14.9⫾2.6 32.5⫾4.7
30.4⫾2.1 69.6⫾2.1
29.5⫾3.4 70.5⫾3.4
35.1⫾4.4 64.9⫾4.4
25.2⫾4.3 75.8⫾4.3
31.3⫾3.8 68.7⫾3.8
⬍0.001
0.844 0.267
0.005 ⬍0.001
0.043 0.430 0.661 0.963
0.026 0.403
a
Percentages may not sum to 100% because of rounding. Rao-Scott 2 tests were used to compare the distribution of characteristics across all four groups. c SCHIP⫽State Children’s Health Insurance Program. b
July 2011 ● Journal of the AMERICAN DIETETIC ASSOCIATION
1023
Table 2. Mean macro- and micronutrient intake reported by low-income, 3- to 5-year-old children in the 1999-2004 National Health and Nutrition Examination Survey by preschool status Preschool Statusa
Overall
Intake
Meanⴞstandard error
Non–Head Start preschool (unweighted nⴝ189)
Head Start (unweighted nⴝ184)
Past preschool (unweighted nⴝ193)
No preschool (unweighted nⴝ384)
4™™™™™™™™™™™™™™™™™™ weighted mean⫾standard error ™™™™™™™™™™™™™™™™™™3 Macronutrient Total energy (kcal) Total protein (g) Total carbohydrates (g) Total fat (g) Saturated fat (g) Cholesterol (mg) Fiber (g) Micronutrient Vitamin A (RAEb) (g) Vitamin B-1 (thiamin) (mg) Vitamin B-2 (riboflavin) (mg) Vitamin B-3 (niacin) (mg) Vitamin B-6 (mg) Vitamin B-9 (folate) (g) Vitamin B-12 (g) Vitamin C (mg) Vitamin E (mg) Calcium (mg) Phosphorous (mg) Magnesium (mg) Iron (mg) Zinc (mg) Copper (mg) Sodium (mg) Potassium (mg) Selenium (mg)
1,720.05⫾33.67 58.65⫾1.18 234.88⫾4.81 63.20⫾1.71 23.24⫾0.70 207.53⫾7.61 11.14⫾0.27
1,668.74⫾58.93 55.51⫾2.37 233.01⫾8.64 59.91⫾2.34 21.28⫾0.89 211.81⫾15.74 11.72⫾0.58
1,710.14⫾56.86 57.54⫾2.23 235.68⫾8.50 62.41⫾2.56 22.62⫾1.09 194.16⫾16.86 10.89⫾0.60
1,657.98⫾58.80 56.13⫾2.39 229.95⫾9.34 59.54⫾2.32 21.83⫾0.76 175.64⫾11.56 10.42⫾0.69
1,780.89⫾60.98 61.89⫾2.03* 238.02⫾8.75 67.00⫾3.30 25.14⫾1.34* 231.84⫾13.18 11.48⫾0.42
501.01⫾13.86 1.38⫾0.03 1.98⫾0.04 14.59⫾0.32 1.51⫾0.03 333.28⫾9.24 4.26⫾0.15 100.78⫾3.94 4.76⫾0.13 911.06⫾25.83 1,096.62⫾23.15 202.59⫾4.23 13.12⫾0.27 9.45⫾0.23 0.88⫾0.02 2,629.50⫾51.08 2,154.04⫾53.52 74.53⫾1.33
468.78⫾22.59 1.33⫾0.06 1.85⫾0.07 15.95⫾0.78 1.49⫾0.07 321.18⫾15.67 4.37⫾0.63 109.45⫾10.73 4.92⫾0.20 785.85⫾43.05 1,010.72⫾42.91 201.07⫾7.30 13.08⫾0.53 9.01⫾0.42 0.87⫾0.03 2,620.06⫾103.40 2,144.78⫾80.76 72.14⫾3.59
480.04⫾29.78 1.34⫾0.05 1.88⫾0.08 16.65⫾0.56 1.46⫾0.07 318.76⫾17.44 4.16⫾0.23 96.51⫾8.68 5.04⫾0.28 875.53⫾54.23 1,088.26⫾37.55 201.62⫾9.19 13.11⫾0.55 9.05⫾0.44 0.85⫾0.04 2,536.63⫾116.91 2,085.32⫾112.95 70.50⫾2.61
501.87⫾30.60 1.41⫾0.07 1.96⫾0.09 16.87⫾0.69 1.50⫾0.09 318.16⫾17.46 4.18⫾0.27 88.65⫾6.90 4.09⫾0.17** 862.16⫾36.49 1,042.47⫾40.83 189.46⫾8.08 12.59⫾0.61 9.17⫾0.49 0.82⫾0.04 2,607.23⫾82.42 2,006.08⫾95.08 70.55⫾2.47
523.97⫾27.24 1.41⫾0.04 2.08⫾0.08* 16.39⫾0.57 1.55⫾0.06 354⫾19.13 4.58⫾0.21 106.87⫾8.30 4.95⫾0.21 1,006.27⫾49.89** 1,165.31⫾39.97** 211.4⫾7.45 13.45⫾0.53 10.00⫾0.40 0.94⫾0.03 2,696.03⫾80.19 2,282.22⫾90.89 79.91⫾2.38a
a
Pairwise t tests used to calculated differences between Head Start and each of the other preschool groups. RAE⫽retinol activity equivalent. *P⬍0.05, for pairwise comparison with Head Start. **P⬍0.01, for pairwise comparison with Head Start. b
Statistical Methods Data were analyzed using the survey procedures in SAS version 9.1 (SAS Institute Inc, Cary, NC) to account for the complex survey design (18). Mean energy, macronutrient, and selected micronutrient intakes were calculated for each preschool group in addition to the percentage of children not meeting the Dietary Reference Intakes for each nutrient. Pairwise comparisons between HS and the other groups were conducted using the SURVEYREG and SURVEYFREQ procedures. Multivariate logistic regression analyses were performed to examine the odds of inadequate nutrient intake for each of the other preschool groups compared with HS. Covariates included age, race/ethnicity, poverty level, teenage pregnancy, household smoking, household size, household receipt of WIC benefits, household reference person foreign born and education level, supplement use, and dietary recall day. This
1024
July 2011 Volume 111 Number 7
study was deemed exempt under federal regulation 45 CFR §46.101(b). RESULTS Description of the Sample Compared with the other groups, HS children were more likely to be non-Hispanic black, to have a mother who was pregnant as a teenager, to live in a household receiving Special Supplemental Nutrition Program for Women, Infants, and Children benefits, and have a foreign-born household reference person. In addition, they were the least likely to be taking dietary supplements (Table 1). Children who were attending a non-HS preschool were more likely to be living in a household in which the household reference person had some college education; conversely, they were less likely to be in households below the federal poverty line and with a smoker present.
Table 3. Percentage of low-income preschool-aged children in the 1999-2004 National Health and Nutrition Examination Survey not meeting dietary guidelines for key nutrients on recall day Preschool Statusb Does not meet guideline for:a
Overall
Head Start (unweighted nⴝ184)
Non–Head Start preschool (unweighted nⴝ189)
Past preschool (unweighted nⴝ193)
No preschool (unweighted nⴝ384)
4™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™ weighted %⫾standard error ™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™3 Macronutrient Protein Total fatc Saturated fatd Micronutrient Vitamin A (RAEe) Vitamin B-1 Vitamin B-2 Vitamin B-3 Vitamin B-6 Vitamin B-9 Vitamin B-12 Vitamin C Vitamin E Calcium Magnesium Iron Zinc Copper Sodiumf Potassium Selenium
2.3⫾0.7 29.3⫾2.1 69.1⫾2.4
4.5⫾2.7 25.2⫾4.3 71.5⫾3.8
0.1⫾0.4*** 30.4⫾4.5 64.8⫾4.8
0.2⫾1.1 26.1⫾3.9 69.3⫾4.2
0.5⫾3.3 32.1⫾4.4 70.3⫾4.1
39.7⫾2.6 5.4⫾1.0 3.7⫾1.0 8.1⫾1.3 8.1⫾1.5 20.3⫾1.8 6.5⫾1.0 12.3⫾2.3 81.0⫾2.1 40.2⫾2.6 15.3⫾2.2 28.8⫾2.8 12.6⫾2.0 7.0⫾1.2 94.8⫾1.0 87.6⫾1.4 3.5⫾0.8
43.3⫾4.1 9.5⫾2.8 8.2⫾2.6 14.6⫾3.7 5.3⫾1.9 20.5⫾4.2 9.4⫾3.5 12.9⫾3.5 79.7⫾2.8 50.4⫾4.3 16.3⫾3.1 26.0⫾4.0 16.5⫾3.5 7.9⫾2.4 96.1⫾1.7 90.8⫾3.2 7.1⫾2.6
45.9⫾5.2 5.0⫾1.9 2.7⫾1.6 4.9⫾2.1*** 7.6⫾2.8 19.0⫾3.5 7.2⫾2.6 14.2⫾4.2 80.0⫾3.9 46.7⫾5.1 17.8⫾4.5 26.8⫾5.4 17.2⫾4.0 8.7⫾3.0 94.9⫾1.9 85.4⫾3.6 1.1⫾0.5***
39.3⫾4.5 2.0⫾1.0*** 2.5⫾1.2* 6.5⫾1.4*** 8.4⫾3.2 28.3⫾4.8 5.6⫾1.7 15.0⫾4.7 89.9⫾2.6*** 42.5⫾4.6 14.1⫾4.0 31.8⫾5.5 10.9⫾2.5 5.8⫾2.3 94.5⫾2.1 94.8⫾2.3 2.3⫾1.0*
35.3⫾4.3 6.1⫾1.8 3.3⫾1.5 8.2⫾2.0 9.3⫾2.2 16.1⫾2.4 5.6⫾1.5 9.3⫾2.3 76.9⫾3.7 31.4⫾3.4*** 14.2⫾3.1 29.3⫾3.5 9.7⫾2.4 6.5⫾1.8 94.7⫾1.6 83.3⫾2.6 4.1⫾1.4
a
Dietary guidelines refer to dietary reference intakes defined as recommended daily allowances or adequate intakes unless otherwise noted. Rao-Scott 2 test used to calculate differences between Head Start and each of the other preschool groups. Guideline for total fat intake defined as percent energy from fat is ⬍40% of total energy intake for 3-year-old children and less than 35% of total energy for 4- to 5-year-old children. d Guideline for saturated fat intake defined as percent energy from saturated fat is ⬍10% of total energy intake for all children. e RAE⫽retinol activity equivalent. f Guideline for sodium intake defined as below the daily tolerable Upper Level for age. *P⬍0.05, for pairwise comparison with Head Start. ***P⬍0.001, for pairwise comparison with Head Start. b c
The age distributions of HS and non-HS preschool children were comparable, whereas past-preschool children tended to be older and no-preschool children younger. The prevalence of overweight did not vary among preschool groups. Mean Dietary Intake Values All four groups reported adequate mean intakes of all key macronutrients and micronutrients except vitamin E, for which all four groups had mean intakes below the agespecific RDA (Table 2). HS children tended to report lower mean intakes of protein, saturated fat, thiamin, riboflavin, niacin, calcium, phosphorous, iron, and selenium than children in the other three categories. The greatest differences in intake were observed between HS and no-preschool children; no-preschool children had significantly higher mean intakes of protein, fat, riboflavin, calcium, and phosphorous. There were no significant differences at the mean level between HS children and non-HS preschool children. No differences were observed at the mean energy level, suggesting that individual
groups were not systematically under- or overreporting intakes. Percentage of Children Not Meeting Recommended Intakes Overall, the nutrients for which the greatest percentage of children did not meet the RDA included folate (20.5%), vitamin A (39.7%), vitamin E (79.7%), calcium (40.2%), iron (28.8%), and potassium (90.8%) (Table 3). In comparison to the other three groups, HS had more children not meeting the requirements for protein, riboflavin, calcium. The prevalence of low niacin intake was 14.6% among HS children, compared to 4.9% in non-HS and 6.5% in pastpreschool children. Fifty percent of HS children did not meet the daily calcium requirements vs only 31.4% of no-preschool children. Many of these differences were statistically significant in pairwise comparisons between HS and other preschool groups (Table 3). Mutivariable Analyses Multivariate analyses showed that children in HS were between 3.1 and 9.6 times more likely to not meet vita-
July 2011 ● Journal of the AMERICAN DIETETIC ASSOCIATION
1025
Table 4. Results of unadjusted and adjusted logistic regression analyses for association between preschool status and not meeting dietary guidelines in the 1999-2004 National Health and Nutrition Examination Surveyab Preschool Status
Intake Macronutrient Protein Unadjusted ORd (95% CIe) Adjusted OR (95% CI) Total fat Unadjusted OR (95% CI) Adjusted OR (95% CI) Saturated fat Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin A (RAEf) Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin B-1 Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin B-2 Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin B-3 Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin B-6 Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin B-9 Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin B-12 Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin C Unadjusted OR (95% CI) Adjusted OR (95% CI) Vitamin E Unadjusted OR (95% CI) Adjusted OR (95% CI) Calcium Unadjusted OR (95% CI) Adjusetd OR (95% CI) Magnesium Unadjusted OR (95% CI) Adjusted OR (95% CI) Iron Unadjusted OR (95% CI) Adjusted OR (95% CI) Zinc Unadjusted OR (95% CI) Adjusted OR (95% CI) Copper Unadjusted OR (95% CI) Adjusted OR (95% CI)
1026
Head Startc (unweighted nⴝ184)
Non–Head Start preschool (unweighted nⴝ189)
Past preschool (unweighted nⴝ193)
No preschool (unweighted nⴝ384)
1.00 1.00
0.08 (0.01, 0.53) 0.07 (0.01, 0.69)
0.24 (0.04, 1.54) 0.20 (0.02, 1.76)
0.72 (0.17, 3.10) 0.64 (0.11, 3.55)
1.00 1.00
1.30 (0.78, 2.16) 1.54 (0.87, 2.74)
1.05 (0.53, 2.06) 1.19 (0.55, 2.54)
1.40 (0.74, 2.67) 2.38 (1.17, 4.85)
1.00 1.00
0.73 (0.46, 1.17) 0.77 (0.46, 1.29)
0.90 (0.50, 1.62) 0.84 (0.45, 1.60)
0.94 (0.54, 1.64) 0.89 (0.50, 1.56)
1.00 1.00
1.11 (0.71, 1.73) 1.24 (0.73, 2.10)
0.85 (0.51, 1.40) 0.89 (0.49, 1.64)
0.71 (0.42, 1.20) 0.89 (0.51, 1.54)
1.00 1.00
0.50 (0.17, 1.47) 0.32 (0.10, 1.00)
0.20 (0.07, 0.53) 0.17 (0.05, 0.57)
0.62 (0.24, 1.63) 0.55 (0.15, 2.07)
1.00 1.00
0.30 (0.08, 1.11) 0.10 (0.03, 0.38)
0.29 (0.08, 1.12) 0.21 (0.05, 0.88)
0.37 (0.12, 1.16) 0.35 (0.11, 1.15)
1.00 1.00
0.30 (0.14, 0.67) 0.24 (0.11, 0.54)
0.41 (0.23, 0.73) 0.31 (0.13, 0.78)
0.53 (0.23, 1.21) 0.36 (0.11, 1.11)
1.00 1.00
1.46 (0.52, 4.13) 1.26 (0.45, 3.57)
1.63 (0.50, 5.30) 1.25 (0.35, 4.55)
1.81 (0.74, 4.46) 1.72 (0.56, 5.23)
1.00 1.00
0.91 (0.43, 1.92) 0.68 (0.32, 1.45)
1.54 (0.75, 3.17) 1.27 (0.61, 2.63)
0.75 (0.40, 1.42) 0.73 (0.36, 1.50)
1.00 1.00
0.74 (0.22, 2.54) 0.61 (0.14, 2.74)
0.57 (0.20, 1.64) 0.57 (0.18, 1.84)
0.57 (0.21, 1.57) 0.68 (0.22, 2.09)
1.00 1.00
1.11 (0.49, 2.53) 1.05 (0.44, 2.50)
1.18 (0.60, 2.33) 0.90 (0.43, 1.91)
0.69 (0.30, 1.57) 0.64 (0.26, 1.54)
1.00 1.00
1.02 (0.56, 1.84) 0.79 (0.46, 1.36)
2.27 (1.34, 3.85) 1.83 (1.02, 3.28)
0.85 (0.53, 1.35) 0.73 (0.42, 1.28)
1.00 1.00
0.86 (0.51, 1.46) 0.92 (0.49, 1.73)
0.73 (0.47, 1.13) 0.79 (0.47, 1.33)
0.45 (0.28, 0.73) 0.55 (0.31, 0.96)
1.00 1.00
1.11 (0.50, 2.50) 1.08 (0.49, 2.39)
0.85 (0.37, 1.91) 0.76 (0.36, 1.58)
0.85 (0.43, 1.71) 1.04 (0.49, 2.22)
1.00 1.00
1.04 (0.56, 1.93) 0.79 (0.39, 1.60)
1.33 (0.73, 2.40) 1.16 (0.59, 2.29)
1.18 (0.67, 2.09) 1.02 (0.50, 2.09)
1.00 1.00
1.05 (0.54, 2.05) 0.77 (0.31, 1.93)
0.62 (0.35, 1.11) 0.48 (0.21, 1.11)
0.54 (0.25, 1.18) 0.51 (0.19, 1.41)
1.00 1.00
1.12 (0.36, 3.52) 0.95 (0.32, 2.84)
0.72 (0.28, 1.86) 0.70 (0.24, 2.02)
0.81 (0.33, 2.03) 0.85 (0.32, 2.32) (continued)
July 2011 Volume 111 Number 7
Table 4. Results of unadjusted and adjusted logistic regression analyses for association between preschool status and not meeting dietary guidelines in the 1999-2004 National Health and Nutrition Examination Surveyab (Continued) Intake
Sodium Unadjusted OR (95% CI) Adjusted OR (95% CI) Potassium Unadjusted OR (95% CI) Adjusted OR (95% CI) Selenium Unadjusted OR (95% CI) Adjusted OR (95% CI)
Preschool Status Head Startc (unweighted nⴝ184)
Non–Head Start preschool (unweighted nⴝ189)
Past preschool (unweighted nⴝ193)
No preschool (unweighted nⴝ384)
1.00 1.00
0.70 (0.36, 1.36) 0.75 (0.37, 1.54)
1.62 (0.84, 3.13) 1.59 (0.75, 3.35)
1.30 (0.74, 2.29) 1.25 (0.67, 2.33)
1.00 1.00
0.59 (0.21, 1.65) 0.54 (0.17, 1.72)
1.85 (0.53, 6.48) 1.93 (0.55, 6.81)
0.51 (0.23, 1.12) 0.72 (0.28, 1.89)
1.00 1.00
0.15 (0.04, 0.54) 0.15 (0.03, 0.74)
0.31 (0.12, 0.82) 0.33 (0.09, 1.13)
0.56 (0.20, 1.57) 0.50 (0.16, 1.58)
a
Dietary guidelines refer to Dietary Reference Intakes. Adjusted for age; race/ethnicity; poverty level; teenage pregnancy; household smoking; household size; household receipt of Special Supplemental Nutrition Program for Women, Infants, and Children benefits; household reference person foreign born and education level; supplement use; and dietary recall day. c Reference category. d OR⫽odds ratio. e CI⫽confidence interval. f RAE⫽retinol activity equivalent. b
mins B-1 to B-3 requirements as compared to non-HS and past-preschoolers (Table 4). Similarly, they were 1.83 times more likely to not meet calcium requirements than no-preschoolers (P⬍0.05). HS children tended to have a higher risk of inadequate intake of protein, saturated fat, thiamin, riboflavin, niacin, vitamin B-12, calcium, magnesium, zinc, copper, and selenium, although in many cases these differences were not significant. Mean Serum Micronutrient Values To help assess the validity of the observed differences in reported dietary intake among the preschool groups, mean levels of all serum micronutrients for which there were available NHANES data were examined (Table 5). Consistent with parental reports, HS children had the lowest mean serum levels for approximately one half of the vitamins and minerals, including folate, vitamin E, selenium, vitamin A, retinol stearate, and gamma tocopherol. Although intake data from one 24-hour recall is not sufficient for direct comparison to serum levels, a poorer serum profile would seem to suggest habitually low intakes and a lower-quality diet on the whole. Thus, the observation that HS children frequently had lower mean serum vitamin levels supports the validity of our findings on reported dietary intake and suggests that observed differences did not result from differential recall and misclassification. DISCUSSION These data demonstrate that, on average, children in HS consume adequate amounts of all key nutrients except vitamin E. However, compared to other low-income children of the same age, HS children appear to be at greater risk for not meeting the RDAs for several vitamins and
minerals, including thiamin, riboflavin, niacin, selenium, and calcium. The high percentage of HS children not meeting calcium requirements (40.2%) is of particular concern given the relationship of dairy consumption to height (19), weight gain (20), and bone health (21,22). Findings in the parent-reported intake data appear to be substantiated by observations at the serum level, which show that HS participants have generally poorer serum vitamin profiles. No differences in the prevalence of overweight among the preschool groups were observed, although children in our low-income sample had a higher percentage of overweight and at risk of overweight than the general population of 2- to 5-year-olds. Previous research on HS nutrition has focused on the quality of school meals only or has used children of nonsimilar socioeconomic backgrounds for comparison. The results of this study add to the literature by assessing intake throughout the day and by comparing micronutrient intake of HS children to that of other children living in poverty. For example, Oakley and colleagues (23) analyzed menus at Child and Adult Care Food Program (CACFP)-funded centers. They found that whereas nearly all menus met the meal-pattern requirements established by the HS Performance Standards (12), following these guidelines did not guarantee consistent nutrition quality of planned meals. In fact, menus from CACFP centers failed to meet one third of the RDAs for energy, vitamins B-12 and E, calcium, iron, and zinc (23). Other studies have also found that average nutrient intakes at school for children in HS or CACFP-funded programs are not consistent with HS standards. In particular, meals served in HS classrooms have been found to be substandard for energy, saturated fat, calcium, vitamin E, sodium, iron, and zinc requirements (24,25). In general, menus and dietary intakes in childcare centers have
July 2011 ● Journal of the AMERICAN DIETETIC ASSOCIATION
1027
Table 5. Adjusted mean⫾standard error (SE) serum micronutrient levels by preschool status in the 1999-2004 National Health and Nutrition Examination Survey Preschool Statusa
Macronutrient Vitamin B-6 (mg) 1 Cohort (n⫽201) Vitamin B-9 (folate) (ng/mL) 3 Cohorts (n⫽736) Folate, RBCb (ng/mL) 3 Cohorts (n⫽762) Vitamin B-12 (g) 3 Cohorts (n⫽737) Vitamin E (g/dL) 2 Cohorts (n⫽448) Iron (g/dL) 3 Cohorts (n⫽749) Selenium (ng/mL) 1 Cohort (n⫽205) Vitamin A (g/dL) 2 Cohorts (n⫽450) Ret stearate (g/dL) 2 Cohorts (n⫽347) Ret palmitate (g/dL) 2 Cohorts (n⫽379) ␥-Tocopherol (g/dL) 2 Cohorts (n⫽413) ␣-Carotene (g/dL) 1 Cohort (n⫽239) Trans beta carotene (g/dL) 1 Cohort (n⫽239) Cis beta carotene (g/dL) 1 Cohort (n⫽239) Vitamin D (ng/mL) 2 Cohorts (n⫽282)
Non–Head Start preschool (unweighted nⴝ189)
Overall
Head Start (unweighted nⴝ184)
Past preschool (unweighted nⴝ193)
No preschool (unweighted nⴝ384)
4™mean⫾SE™3
4™™™™™™™™™™™™™™™™™™™™™™ weighted mean⫾SE ™™™™™™™™™™™™™™™™™™™™™™™ 3
44.77⫾4.20
46.68⫾4.16
50.97⫾9.83
49.95⫾8.35
38.01⫾3.95
17.99⫾0.52
18.35⫾0.70
18.07⫾0.86
18.71⫾0.96
17.36⫾0.84
283.76⫾5.83
268.84⫾6.24
297.26⫾15.22
280.39⫾12.61
284.64⫾5.99*
815.87⫾21.12
874.76⫾48.83
789.40⫾32.59
816.24⫾33.48
805.38⫾33.27
799.01⫾16.00
730.98⫾20.14
854.14⫾25.75**
809.31⫾44.85
790.29⫾20.24*
74.36⫾1.38
72.77⫾4.37
77.37⫾3.07
72.51⫾2.97
74.42⫾2.32
106.99⫾1.27
101.25⫾2.88
108.05⫾2.15*
113.04⫾2.89*
106.37⫾1.46
33.61⫾0.50
31.86⫾1.17
35.19⫾0.83*
33.37⫾0.97
33.51⫾0.58
0.56⫾0.07
0.44⫾0.04
0.68⫾0.23
0.46⫾0.04
0.60⫾0.07*
2.28⫾0.18
2.12⫾0.18
2.78⫾0.56
1.95⫾0.17
2.24⫾0.21
205.35⫾14.20
233.64⫾10.97
207.49⫾28.67
2.84⫾0.31
3.52⫾0.81
2.95⫾0.55
3.50⫾0.75
2.26⫾0.16
14.09⫾0.69
14.78⫾1.37
16.30⫾1.36
14.49⫾2.22
12.68⫾0.89
0.70⫾0.03
0.73⫾0.06
0.80⫾0.06
0.71⫾0.09
0.65⫾0.04
28.76⫾0.86
28.34⫾1.67
27.26⫾1.07
29.90⫾1.19
28.74⫾1.19
216.67⫾8.25
216.53⫾7.89
a
Pairwise t tests used to calculated differences between Head Start and each of the other preschool groups. RBC⫽red blood cell. *P⬍0.05, for pairwise comparison with Head Start. **P⬍0.01, for pairwise comparison with Head Start. b
been shown to be lacking in energy, iron, niacin, thiamin, calcium, vitamin B-6, vitamin A, folic acid, and protein (26-28). Worobey and colleagues (29) found that low-income children attending an HS preschool consumed similar levels of protein, carbohydrate, and fat as middleincome children at a private preschool in the same area. However, they also found that the home diets of the HS children were higher in fat and carbohydrates. Thus, studies to date have reported mixed results on the quality of meals served in Head Start and preschool programs. This discrepancy is likely attributable to differences in the data sources, dietary measurement techniques, and comparison groups. This study is the first to compare the diets of HS children to those of other low-income preschool-aged children and to show that children in HS are at higher risk for not
1028
July 2011 Volume 111 Number 7
meeting the RDAs for several vitamins and minerals. Although it could not be determined whether these findings are the consequence of suboptimal nutrition at school or at home, the results of this study present opportunities for intervention. If the observed differences in diet quality at the preschool level are a result of the food served at HS programs, then stronger HS policies, increased CACFP funding, and additional staff training may help to improve the nutritional intake of children in HS. Although HS centers consult dietetics practitioners more frequently than non-HS centers when designing menus (30), these dietetics practitioners may not have the capacity to offer healthier choices without stronger policies or expanded budgets. If, on the other hand, the differences are a result of food served at home, then greater communication between HS centers and parents
in the form of nutrition awareness and education may improve nutritional intake (31,32). The observation that HS children were more likely to have foreign-born heads of household also warrants discussion. Households with foreign-born heads may have certain cultural or religious practices that influence diet at home, and foreign-born parents may have different beliefs or expectations from government programs like HS. In a survey of obesity prevention in HS, 58% of HS directors cited cultural beliefs about food as a barrier to healthy eating by HS children (33). Although diets in these households may be healthy and meet daily requirements, preschool educators may find it challenging to educate these children and families about nutrition practices and quality. Given this observation, nutrition information relayed to children and parents should be mindful of differences in cultural practices and foods while emphasizing the importance of adhering to a healthy diet. This study illustrates important differences in diet quality by preschool status, but study limitations must be acknowledged. First, a single 24-hour recall was used to evaluate preschoolers’ dietary intakes. Although a single recall has been shown to be sufficient for calculating mean food and nutrient intakes on a population level, the use of multiple recalls is generally preferred for estimating the prevalence of high or low intakes (34). The absence of multiple recall days for all members of the NHANES sample prevented us from being able to estimate usual intake. However, this limitation would not have been expected to cause differential biases between preschool groups because the same methods of calculating inadequate intake were used for all groups. Second, the use of parents as proxies for the measurement of preschoolers’ dietary intake may increase the likelihood of misreporting. Parents with children in preschool or child care may be less informed about the types and quantities of food their child consumes outside of the home. Whereas the possibility of differential misreporting exists, all preschool groups had similar mean caloric intakes with relatively narrow standard errors, suggesting that no group had systematically under- or overreported children’s intakes. Third, variability in the nutrition environment and diet quality within the category of non-HS preschools may be sizeable, such that treating these schools as a unit may not be appropriate. In fact, this limitation may explain why the fewest number of differences were observed between this group and HS children. In addition, children classified as past preschoolers may have attended either Head Start or non-HS programs. Fourth, students who attended HS part day vs full day could not be differentiated nor could foods consumed at home vs school. Fifth, observed dietary intakes could not be directly related to serum levels because data on most serum vitamins were not available and because usual intakes could not be estimated. Sixth, a greater percentage of HS students lived in households with incomes below the federal poverty level, which may influence access to healthy foods at home. Finally, the problem of multiple comparisons may have been present in the results. To help distinguish highly significant findings from weaker ones, separate indicators for varying levels of significance are indicated.
CONCLUSIONS Children living in poverty are at high risk for nutritional deficiencies, obesity, and subsequent inequities in health (ie, hypertension, metabolic syndrome, and diabetes) (6,8). Despite federal efforts to improve nutrition in lowincome children through the HS program, the diets of HS preschoolers do not appear to be on par with those of other preschool-aged children; in general, HS children appear to have poorer quality diets than non-HS children of the same socioeconomic status. Because nutrition provides an important foundation for academic performance and overall health, ensuring that HS children receive a diet that is as equally nutritious as that of other children is imperative for success. Further research is needed to delineate the source of the observed differences in diet quality and ultimately to design strategies for enhancing the diets of HS children both at home and in school. STATEMENT OF POTENTIAL CONFLICT OF INTEREST: No potential conflict of interest was reported by the authors. FUNDING/SUPPORT: This project was independently performed. The authors have no funding sources to disclose. References 1. Polhamus B, Dakenius K, Thompson D, Scanlon K, Borland E, Smith B, Grummer-Strawn L. Pediatric Nutrition Surveillance 2003 Report. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2004. 2. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States, 19992004. JAMA. 2006;295:1549-1555. 3. Alaimo K, Olson CM, Frongillo EA, Briefel RR. Food insufficiency, family income, and health in US preschool and school-aged children. Am J Public Health. 2001;91:781-786. 4. Kotch J, Shackelford J. The nutritional status of low-income preschool children in the United States: A review of the literature. 1989. Food Research and Action Center Web site. http://frac.org. Accessed April 4, 2009. 5. Rose D, Smallwood D, Blaylock J. Socioeconomic factors associated with the iron intake of preschoolers in the United States. Nutr Res. 1995;15:1298-1309. 6. Birch LL. Development of food preferences. Annu Rev Nutr. 1999;19: 41-62. 7. Birch LL. Effects of peer models’ food choices and eating behaviors on preschoolers’ food preferences. Child Dev. 1980;51:489-496. 8. Skinner JD, Carruth BR, Bunds W, Ziegler PJ. Children’s food preferences: A longitudinal analysis. J Am Diet Assoc. 2002;102:16381647. 9. Hoerr S, Utech AE, Ruth E. Child control of food choices in Head Start families. J Nutr Educ Behav. 2005;37:185-190. 10. Capizzano J, Adams G. The Hours That Children Under Five Spend in Child Care: Variation Across States. Washington, DC: Urban Institute; 2000. 11. Wright JD, Borrud LG, McDowell MA, Wang C, Radimer K, Johnson CL. Nutrition assessment in the National Health and Nutrition Examination Survey 1999-2002. J Am Diet Assoc. 2007;107:822-829. 12. Head Start Program performance standards and other regulations: Part 1304.23: Child nutrition. Office of Head Start Web site. http:// eclkc.ohs.acf.hhs.gov/hslc. Accessed April 10, 2009. 13. National Health and Nutrition Examination Survey. National Center for Health Statistics, Centers for Disease Control and Prevention Web site. www.cdc.gov/nchs/nhanes.htm. Accessed April 7, 2009. 14. USDA Food and Nutrient Database for Dietary Studies. US Department of Agriculture Web site. http://www.usda.gov/wps/portal/usda/ usdahome. Accessed April 7, 2009. 15. Kleinman RE. Pediatric Nutrition Handbook. 6th ed. Elk Grove Village, IL: American Academy of Pediatrics; 2009:1294-1295. 16. Dietary Guidelines for Americans 2005. US Department of Agricul-
July 2011 ● Journal of the AMERICAN DIETETIC ASSOCIATION
1029
17. 18.
19. 20. 21. 22. 23. 24. 25.
ture Web site. http://www.health.gov/dietaryguidelines. Accessed July 6, 2009. A SAS Program for the CDC Growth Charts. Atlanta, GA: Centers for Disease Control and Prevention; 2005. Gossett JM, Jo CH, Simpson PM. US Health and Nutrition: SAS survey procedures and NHANES. Paper presented at 31st Annual SAS Users Group International Conference, March 26-29, 2006; San Francisco, CA. http://www2.sas.com/proceedings/sugi31/140-31.pdf. Accessed April 7, 2009. Wiley AS. Does milk make children grow? Relationships between milk consumption and height in NHANES 1999-2002, Am J Hum Biol. 2005;17:425-441. Moore LL, Bradlee ML, Gao D, Singer MR. Low dairy intake in early childhood predicts excess body fat gain. Obes Res. 2006;14:1010-1018. Black RE, Williams SM, Jones IE, Goulding A. Children who avoid drinking cow’s milk have lower dietary calcium intakes and poor bone health. Am J Clin Nutr. 2002;76:675-680. Teegarden D, Lyle RM, Proulx WR, Johnston CC. Previous milk consumption is associated with greater bone density in young women. Am J Clin Nutr. 1999;69:1014-1017. Oakley CB, Bomba AK, Knight KB, Byrd SH. Evaluation of menus planned in Mississippi child-care centers participating in the Child and Adult Care Food Program. J Am Diet Assoc. 1995;95:765-768. Fox MK, Glantz FB, Geitz L, Burstein N. Early Childhood and Child Care Study: Nutritional Assessment of the CACFP. Cambridge, MA: Abt Associates Inc; 1997. Bollella MC, Spark A, Boccia LA, Nicklas TA, Pittman BP, Williams
1030
July 2011 Volume 111 Number 7
26. 27. 28. 29. 30.
31. 32.
33. 34.
CL. Nutrient intake of Head Start children: Home vs school. J Am Coll Nutr. 1999;18:108-114. Briley M, Buller A, Roberts-Gray CC, Sparkman A. What is on the menu at the childcare center? J Am Diet Assoc. 1989;89:771-774. Briley M, Roberts-Gray C, Rowe S. What can children learn from the menu at the child-care center? J Community Health. 1993;18:363-377. Padget A, Briley ME. Dietary intakes at child-care centers in central Texas fail to meet Food Guide Pyramid recommendations. J Am Diet Assoc. 2005;105:790-793. Worobey J, Worobey HS, Adler AL. Diet, activity, and BMI in preschool-aged children: Differences across settings. Ecol Food Nutr. 2005;44:455-466. Gupta RS, Pascoe JM, Blanchard TC, Langkamp D, Duncan PM, Gorski PA, Southward LH. Child health in child care: A multi-state survey of Head Start and non-Head Start child care directors. J Pediatr Health Care. 2009;23:143-149. Briley ME, Jastrow S, Vickers J, Roberts-Gray C. Dietary intake at child-care centers and away: Are parents and care providers working as partners or at cross-purposes? J Am Diet Assoc. 1999;99:950-954. Spurrier NJ, Magarey AA, Golley R, Curnow F, Sawyer MG. Relationships between the home environment and physical activity and dietary patterns of preschool children: A cross-sectional study. Int J Behav Nutr Phys Act. 2008;5:31-42. Hughes CC, Gooze RA, Finkelstein DM, Whitaker RC. Barriers to obesity prevention in Head Start. Health Affairs. 2010;29:454-462. Willett W. Nutritional Epidemiology, 2nd ed. New York, NY: Oxford University Press; 1998:54.