Nutritional Quality of Meals Compared to Snacks in Child Care

Abstract

Background

Most young children are in child care. Previous studies suggest that children may receive insufficient vegetables, and foods and beverages with added sugars, fats, and sodium in these settings. None have compared the nutritional quality of meals to snacks.

Methods

Directors from 258 full-day child-care centers in two urban counties of southwestern Ohio were surveyed via telephone in the fall of 2009 about their nutrition practices, and asked to provide a current menu. Lunch and afternoon snack menus were categorized according to average weekly frequency for fruits, vegetables, lean meats, juice (100%), and sweet or salty foods served. Frequencies were compared by meal occasion (lunch vs. snack) using the Fisher exact test.

Results

Most (60%) directors reported serving 2% milk to children ≥3 years; 31% served whole milk. Menu analysis demonstrated the composition of lunches differed from snacks (p<0.0001) in all food categories. A total of 87% centers rarely (<1 time per week) listed nonstarchy vegetables for snacks, but 67% of centers included them at lunch ≥3 times per week. Juice (100%) was on snack menus >2 times per week in 37% centers, but in only 1 center as a regular component of lunch. Similarly, 87% centers listed sweet and salty foods at snack ≥3 times per week, but rarely at lunch.

Conclusions

Despite efforts to improve children's diets in child care, meals—and particularly snacks—still lack whole fruits and nonstarchy vegetables and contain added sugars and fats. Snacks represent a missed opportunity to improve the nutritional quality of foods served in childcare.

Introduction

Child care is an important setting for promoting healthy eating in children.^1,2 The majority of US children attend child care prior to entering school^3,4 and spend long hours in these settings daily.^5,6 Among children 2–5 years old, the combined prevalence of overweight and obesity is 27%.⁷ Research suggests that child-care providers play an important role in shaping young children's eating habits and food preferences.^8,9 Furthermore, the dietary intake of other children in the same classroom may influence a child's consumption through a peer effect,¹⁰ particularly as it relates to trying new foods¹¹ such as vegetables.^12,13 Thus, both researchers and health professionals have targeted child-care programs in an effort to curb childhood obesity.^1,2,14

Previous studies have examined the nutritional quality of foods and beverages served in child-care programs. Overall, these studies found that children were served insufficient amounts of vegetables,^15–18 grains^15,17–19 (especially whole grains^15,18), and foods rich in iron,^16,18–21 zinc,^18–20 and niacin.^16,19–21 A few studies have also reported inadequate servings of fruits,^15,17 meats,^15,17 and fiber.^19,22 Conversely, most studies have reported that foods served at meals and snacks are high in fat^18–22 (e.g., full-fat milk,¹⁵ high-fat meat,^15,16,21 or fried foods¹⁵), added sugars^{15,16,18,20,21} (e.g., cookies/pastries and sweetened beverages^15,16), and sodium.^18,19,21,22 Some of these studies were published over a decade ago,^16,18–21 thus predating the current obesity epidemic²³ and multiple public health efforts²⁴ to improve the nutritional quality of foods served in child-care settings. As a result, it is unclear if these patterns persist. Moreover, most of these studies^15–20,25 have examined child-care programs in the southern United States, so it is unknown to what extent food servings in other parts of the country mirror the patterns noted above. Last, none of these studies was stratified by type of meal occasion (meal vs. snack) to compare the nutritional quality. In our own comparison of menus to actual foods and beverages served in 84 child-care centers in North Carolina,²⁶ juice (100%) and foods of low nutritional value (e.g., cookies, chips) were found to be the most common items served at afternoon snack, while they were the least common foods served at lunch. Similarly, vegetables and proteins were often served at lunch, but rarely at afternoon snack.

A few studies have examined the food environment within child-care settings, which may include whether or not child-care providers sit at the table and eat with children, adult–child interactions during meals and snacks, and nutrition-related training for providers. Staff sitting with children at meals has been reported to be common practice in some child-care settings^27,28 but not others.^27,29 Similarly, the extent to which child-care providers encourage—but not coerce—children to try new foods is highly variable.^25,27–29 These studies highlight the need for additional information describing the food environment and caregivers' nutrition practices in order to focus intervention efforts. The purpose of this study was to compare the nutritional quality of meals and snacks in full-day child-care centers in southwestern Ohio and to describe nutrition-related practices in these settings.

Methods

Overview and Setting

The data used for this analysis were collected as part of the follow-up assessments for a 6-month population-based intervention conducted in the summer of 2009. The target population for this state-sponsored intervention, which was based on the Nutrition and Physical Activity Self-Assessment for Child Care (NAP SACC) intervention,²⁴ was all licensed full-day child-care centers in Montgomery County, Ohio (Dayton area). Hamilton County (Cincinnati area) served as the control county. Follow-up assessments showed no measurable intervention effect for self-reported nutrition or physical activity policies and practices,³⁰ thus control and intervention counties were combined for this analysis.

Sample and Recruitment

Eligible participants for the baseline assessment included all directors or senior administrative staff from licensed full-day child-care centers serving children 3–6 years of age in each county (n=185 in Hamilton County, n=151 in Montgomery County) according to a list obtained from the local child-care resource and referral agency in 2008. Centers were mailed a letter of introduction, and study staff attempted to reach the center up to eight times at different times of the day to schedule an interview before the center was considered unreachable. Only the centers participating in the baseline assessment (August, 2008–March, 2009) were eligible to participate in follow-up assessment (September–December, 2009). Interviews were conducted over the telephone by trained research staff, lasted an average of 25 minutes, and responses were directly recorded into a computerized database. Centers received a $15 gift card for completing each assessment. During the follow-up assessment period only, centers from both counties were asked to mail, fax, or e-mail a lunch and snack menu from the preceding month. Locally, centers vary in morning meal occasion offerings (breakfast or morning snack or both), but all full-day centers offer lunch and a single mid-afternoon snack; thus, only these menus were used in this analysis. The protocol was reviewed and determined to be exempt by the Cincinnati Children's Hospital institutional review board.

Measures

The instrument used for this survey was the Early Learning Environments for Physical Activity and Nutrition Telephone Survey (ELEPhANTS), which included measures of the nutrition, physical activity, and TV/media environment of child-care centers. The development and refinement of this instrument has been discussed elsewhere,^31,32 but briefly the instrument included items and response scales from the 2004 and 2007³³ versions of the NAP SACC self-assessment tool²⁴ that have previously been shown to have good reliability and validity.³⁴ Items that had low agreement with objectively observed behaviors were modified and reviewed for content validity by registered dietitians, as well as experts in early education, public health, nutritional epidemiology, and scale development. Furthermore, the instrument was assessed for face validity prior to administration through cognitive interviews with 13 child-care providers.

The instrument focused on nutrition practices that could not be determined from reviewing a menu, e.g., the percent milk fat typically served and how the meals were served (staff served or children served themselves). Last, because menus have been shown to be a somewhat unreliable source of information about foods such as cookies and crackers,^35,36 directors were also asked how often sweet foods such as cookies, animal crackers, fruit gummy snacks, pastries, donuts, muffins, and other sweets and salty foods such as crackers, pretzels, chips, popcorn, and cheese puffs were served in their centers.

Analysis

SAS^®, version 9.3 (SAS Institute, Cary, NC) was used for data management and analysis. Demographics of centers providing a menu were compared with centers not providing a menu using a chi-squared or Fisher exact test, as appropriate, for categorical variables and the Wilcoxon rank sum for continuous measures. The average weekly frequency of 10 food group categories, derived from NAP SACC,²⁴ was tallied by meal occasion by a nutritionist (A.H.) for lunch menus (n=150) and snack menus (n=143) separately. Because the number of days of menus collected from each center varied, weekly averages of each food group served were computed. Tallied frequencies were then categorized into one of four response categories for each food group item, also derived from NAP SACC.²⁴ The frequency of food items served at lunch versus snack was compared using the Fisher exact test. A post hoc analysis was also conducted to examine the trends in food group item frequency categories of NAP SACC comparing meals to snacks. The Cochran–Armitage Trend test was used, but because of the very small cell sizes for some of the response categories, the analysis involved combining the four response categories into two or three response categories.

Results

Telephone surveys were completed with 258 out of the 336 centers potentially eligible at the start of the intervention (80% response rate; n=152 in Hamilton County, n=106 in Montgomery County). The final convenience sample (Table 1, column 3) included a diversity of for-profit status (44% not-for profit) and affiliations (20% Head Start, 45% independent), as well as enrolled age groups, race, and socioeconomic status of children served. Of the 258 centers, 170 (55%, Table 1, column 2) centers provided a lunch or afternoon snack menu; 123 centers provided both. There were no statistically significant differences in the demographics of centers that supplied a menu compared to those that did not, except that a greater proportion of those not supplying a menu were part of a national chain (24% vs. 14%, p=0.04) (Table 1). On the telephone survey, 61% of 258 directors indicated that center lunches were prepared by on-site cooks, 25% were prepared by an off-site catering company, 8% by parents, and 5% by a combination of parents and professional cooks. For snacks, the percentages were similar (66%, 26%, 3%, and 5%, respectively). Among the 170 centers that provided a lunch (n=150) or snack (n=143) menu (Table 2), centers submitted a mean (standard deviation) 18 (6) days for lunch, and 17 (7) days for afternoon snack.

Table 1.

Demographics of Participating Child-Care Centers as Reported by Center Director

Center characteristic	No menu provided n=88	Menu provided n=170	Total n=258	p valuea
Enrollment by age-group, median (Interquartile range)
0–18 months	9 (0, 17)	9 (0, 16)	9 (0, 16)	0.58†
18–36 months	12 (7, 20)	13 (7, 21)	12 (7, 20)	0.62†
3–5 years	22 (14, 42)	25 (14, 37)	25 (14, 40)	0.94†
For profit, n (%)b	53 (61)	90 (53)	143 (56)	0.22*
Not for profit, n (%)b	34 (39)	80 (47)	114 (44)
Affiliations (choose all that apply), n (%)
Local chain	16 (18)	28 (16)	44 (17)	0.73*
National chain	21 (24)	23 (14)	44 (17)	0.04*
Church-based	12 (14)	34 (20)	46 (18)	0.21*
Worksite-based	2 (2)	3 (2)	5 (2)	1.00**
College-affiliated	1 (1)	6 (4)	7 (3)	0.43**
Government-affiliated	2 (2)	5 (3)	7 (3)	1.00**
Hospital-based	2 (2)	4 (2)	6 (2)	1.00**
Independent	38 (43)	79 (46)	117 (45)	0.61*
Head Start	15 (17)	36 (21)	51 (20)	0.47*
Participates in Child and Adult Care Food Program (CACFP), yes, n (%)	48 (61)	89 (57)	137 (58)	0.59*
Percent of children participating in CACFP, median (25th %ile, 75th %ile)	50 (0, 90)	28 (0, 95)	40 (0, 92)	0.81††
Tuition, median (25th %ile, 75th %ile)	156 (145, 175)	159 (145, 175)	159 (145, 175)	0.31†
Percent of children on subsidized tuition, median (25th %ile, 75th %ile)	80 (30, 95)	75 (16, 95)	75 (25, 95)	0.53†
bPrimary race (race of >50% of current enrollees), n (%)
White, non-Hispanic	34 (39)	74 (44)	108 (43)	0.85**
Black, non-Hispanic	32 (37)	59 (35)	91 (36)
Mixed race/biracial	1 (1)	2 (1)	3 (1)
No primary race	20 (23)	32 (19)	52 (20)
Any staff trained in nutrition, yes, n (%)	57 (66)	107 (63)	164 (64)	0.60*

^aP value of the corresponding test (^*Chi-square ^**Fisher exact ^†t-test ^††Wilcoxon rank sum) comparing frequency of the characteristic among centers providing menu (n=170) vs. those that did not (n=88).

^bTotal n for these items does not add up to 258 centers due to missing data: 1 center declined to provide for-profit status, and 4 centers declined to provide an assessment of the racial/ethnic makeup of families served.

Table 2.

Composition of Lunch versus Afternoon Snack: Results from a Menu Analysis and Telephone Survey

	Lunch n=150 menus n (%)	Afternoon snack n=143 menus n (%)	p value
Fruit (whole, canned, or frozen, excluding 100% juice) is offereda			<0.0001†
3 times per week or less	11 (7)	130 (91)
4 times per week	14 (9)	3 (2)
Once per meal occasion	124 (83)	10 (7)
Twice or more per meal occasion	1 (1)	0 (0)
Vegetables (not including french fries, Tater Tots, hash browns, or dried beans) are offered			<0.0001†
2 times per week or less	4 (3)	141 (99)
3–4 times per week	42 (28)	2 (1)
Once per meal occasion	103 (69)	0 (0)
Twice or more per meal occasion	1 (<1)	0 (0)
Vegetables other than potatoes, corn, and green beans are offered			<0.0001†
Less than once per week	2 (1)	125 (87)
1–2 times per week	47 (31)	16 (11)
3–4 times per week	95 (63)	2 (1)
Once or more per meal occasion	6 (4)	0 (0)
Fried or prefried potatoes are offered			<0.0001†
Once or more per meal occasion	0 (0)	0 (0)
3–4 times per week	1 (1)	0 (0)
1–2 times per week	26 (17)	1 (1)
Less than once per week or never	123 (82)	142 (99)
Fried or prefried meats or fish are offered			<0.0001†
Once or more per meal occasion	0 (0)	0 (0)
3–4 times per week	1 (1)	0 (0)
1–2 times per week	59 (39)	0 (0)
Less than once per week or never	90 (60)	143 (100)
High-fat meats (sausage, bacon, hot dogs, bologna, ground beef) are offered			<0.0001†
Once or more per meal occasion	0 (0)	0 (0)
3–4 times per week	17 (11)	0 (0)
1–2 times per week	120 (80)	0 (0)
Less than once per week or never	13 (9)	143 (100)
Beans or lean meats (baked or broiled turkey, chicken, or fish) are offered			<0.0001†
Less than once per week	15 (10)	142 (99)
1–2 times per week	96 (64)	1 (1)
3–4 times per week	38 (25)	0 (0)
Once or more per meal occasion	1 (1)	0 (0)
High-fiber, whole-grain foods (whole wheat bread, oatmeal, brown rice) are offered			0.03†
Once per week or less	127 (85)	132 (92)
2–4 times per week	23 (15)	10 (7)
Once per meal occasion	0 (0)	1 (1)
Twice or more per meal occasion	0 (0)	0 (0)
100% fruit juice offered on the menu			<0.0001†
Twice or more per meal occasion	0 (0)	7 (5)
Once per meal occasion	1 (1)	18 (12)
3–4 times per week	0 (0)	28 (20)
2 times per week or less	149 (99)	90 (63)
Sweet or salty foods are offered			<0.0001†
Once or more per meal occasion	0 (0)	55 (38)
3–4 times per week	0 (0)	70 (49)
1–2 times per week	20 (13)	16 (11)
Less than once per week or never	130 (87)	2 (1)
bSurvey data (n=258 centers, total number of responses indicated by question/meal occasion)
Cookies, animal crackers, fruit gummy snacks, donuts, muffins etc. served as part of a….	(n=254)	(n=255)	<0.0001*
Once or more per meal occasion	0 (0)	9 (4)
3–4 times per week	2 (1)	21 (8)
1–2 times per week	17 (7)	120 (47)
Rarely, less than once per week	30 (12)	63 (25)
Never	205 (80)	42 (16)
Crackers, pretzels, popcorn, potato chips, corn or tortilla chips, cheese puffs, etc. served as part of a…	(n=256)	(n=255)	<0.0001*
Once or more per meal occasion	1 (<1)	16 (6)
3–4 times per week	1 (<1)	26 (10)
1–2 times per week	23 (9)	123 (48)
Rarely, less than once per week	43 (17)	69 (27)
Never	188 (73)	21 (8)

^aItems and response categories are from the Nutrition and Physical Activity Self-Assessment for Child Care (NAP SACC).³³ Order of response categories is from least healthy to most healthy (best practice).

^bItems and response categories are from the Early Learning Environments for Physical Activity and Nutrition Telephone Survey (ELEPhANTS).^31,32

^*Chi-squared test.

^†Fisher's exact test.

The composition of meals and snacks on these menus is detailed in Table 2. In all food categories, the composition of lunches was significantly different from the composition of snacks. For instance, a fruit (including whole, canned, frozen, but excluding 100% juice) was a component of snack less than four times per week in 91% of centers, but 93% of centers listed a fruit at lunch at least four times per week. Nonstarchy vegetables were listed as being served at snacks less than once per week at 87% of centers, but were listed as being a component of lunches at least three times per week in 67% of centers, and one to two times per week at 31% of centers. Similarly, beans or a lean meat (excluding breaded meats and high-fat meats) were listed as a component of snacks by only one center, but 90% of centers included beans or lean meats as a component of lunches at least once per week. Fruit juice (100%) was listed as a component of snack more than twice per week at 37% of centers, but only one center listed juice as a regular component of lunch.

Sugar-sweetened beverages (e.g., Kool-aid) were not listed on any lunch or snack menu; in response to the telephone survey, 246 centers (95%) reported never serving sweetened beverages. Sweet or salty foods were rarely listed as a component of lunch (87% of centers listed these items less than once per week or never on lunch menus), but were listed as a daily component of snack in 38% of centers, and a component of three to four snacks per week in an additional 49% of centers. When broken down by type (sweet or salty), directors on surveys indicated that both of these items were rarely or never served at lunch (90% of centers), but sweets were served at least once a week at snack in 59% of centers and salty foods were served at least once per week at snack in 64% of centers. The post hoc analysis to eliminate small cell sizes also showed significant differences in the composition of meals and snacks for each of the 12 food categories listed in Table 2 (p<0.0001 for all items except fiber/whole grain, p=0.04; data not shown).

Looking at nutrition practices (Table 3), 179 (69%) centers reported that drinking water was easily visible and available for self-service in the classrooms. Reduced-fat (2%) milk was reported to be served in 45% of centers, whole milk in 31% of centers, the recommended³⁷ 1% or skim milk in 7% of centers. An additional 15% of centers that also served toddlers ages 18–36 months reported serving 2% milk to children 3 years and older, but whole milk to children under age 3, who are often grouped with children under age 2 and for whom whole milk is recommended. Most directors (84%) reported that at least some staff sit with children during meals and consume the same food as the children always (48%) or most of the time (22%). For meal service, directors reported that staff portioned the food on children's plates in 51% of centers, and that children serve themselves in 37% of centers. In the remaining centers, food was prepared by parents (lunch box). When children were hesitant about trying new or less favorite foods, 83% of directors indicated that staff would encourage the children to try them. When children requested seconds, directors indicated that in 78% of centers children were served whatever they asked for; directors of only three centers indicated that staff modeled best practices by helping children determine if they were still hungry before serving seconds. Regarding nutrition training, 164 (64%) of directors indicated at least one staff member had undergone specialized training in early childhood nutrition, most often provided by the USDA (37%), as part of their college or associates degree (21%), through the local child-care resource and referral agency (14%), or as a part of an in-service or continuing education (11%). In 79 of the centers (48% of centers with any nutrition training, or 31% of all centers), this training was undertaken on an annual basis; in 49 centers this training was undertaken once.

Table 3.

Nutrition Practices Reported by Center Directors in a Telephone Survey

	N	n (%)
How do children receive drinking water indoors?	258
Easily visible and available for self service		179 (69)
Easily visible and available on request		73 (28)
Available during designated water breaks		5 (2)
Not freely available		1 (<1)
What type of milk is usually served to children ages 2 and older?	258
Skim or nonfat		4 (1)
1% low fat		15 (6)
2% reduced fat		115 (45)
2% for children 3 years and older; whole for those under 3 yearsa		40 (15)
Whole or regular milk		79 (31)
Don't know		5 (2)
How often do children and (at least some) staff sit down together for meals?	256
Always		215 (84)
Most of the time		17 (7)
Some of the time		9 (3)
Rarely or never		14 (5)
Don't know		1 (<1)
How often do (at least some) staff consume the same food and drinks as the children?	258
Always		123 (48)
Most of the time		56 (22)
Some of the time		46 (18)
Rarely or never		32 (12)
Don't know		1 (<1)
How is the food for meals served?	258
Children serve themselves		97 (37.5)
Staff portion the meals onto plates		133 (51.5)
Lunch box (parents provide meals)		28 (11)
How does center handle children who do not like or say they do not like the foods they are served (‘picky eaters’)?	257
Children are encouraged to try new or less favorite food		214 (83)
Children offered alternate food not on the menu		21 (8)
Children allowed to eat what they want (not encouraged to try food, not offered an alternative)		17 (7)
Other		4 (1.6)
Don't know		1 (<1)
What does staff do when child asks for seconds (assuming there is plenty of food)?	256
Child is served whatever they ask for		200 (78)
Child is encouraged to try everything on his plate before getting seconds		29 (11)
Child must finish everything on his plate before getting seconds		8 (3)
Staff tries to help child determine if he is still hungry before serving seconds		3 (1)
Never give seconds		16 (6)
Any staff member in center with specialized training in preschool-age nutrition, feeding or menu-planning? Yes, n (%)	256	164 (64)
Percent of staff with any training in nutrition? %, median (IQR)	164	37 (14–98)
How often does this staff undergo training in nutrition?	164
Annually or more than once a year		79 (48)
Once every 2–3 years		24 (15)
Infrequently (every 4 years or greater)		9 (5)
One time (e.g., as part of education)		49 (30)
Don't know		3 (2)

^aTo be eligible to participate in this study, centers had to serve children 3 years and older, but 229 centers also currently served at least 1 child under the age of 3 years.

IQR, interquartile range.

Discussion

To our knowledge, this is the first study of the nutritional composition of lunches compared to afternoon snacks in full-day child-care centers. For all food categories, the types of foods and beverages listed as being served at snack differed significantly from those at lunch. Fruits, vegetables, and meats/beans were rarely included as a component of snack but were listed almost daily as a component of lunches. Conversely, 100% fruit juice was listed as a component of snack at least three times per week in over one-third of centers, but in only a single center as a regular component of lunch. Similarly, 87% of center menus listed sweet and salty foods, such as animal crackers, fruit gummy snacks, pretzels, and crackers, as a component of snack at least three times a week, but rarely or never as a component of lunch.

The frequency of whole grains, fruits, vegetables, and high-fat, high-sugar, and high-salt foods reported on the menus in this sample of centers from southwestern Ohio generally mirrors the patterns reported in the southeastern United States.^15–20,25 Fruits and vegetables other than fried potatoes were listed as a daily component of lunches for most centers, which is consistent with a recent 1-day observation of intake among 96 child-care centers in North Carolina.²⁵ Prefried potatoes, prefried meats, high-fat meats, and whole grains were uncommonly listed on menus as a component of lunches and afternoon snacks, also consistent with recently observed practices in child-care centers.^15,18,25 High-sugar/high-salt foods and 100% juice, however, were listed as common components of snacks, as others have reported.^15,16,25 Fewer centers (31%) reported serving whole milk to preschool-aged children compared to recent studies^15,25 in North Carolina (53–58%). More centers (69%) reported that drinking water was freely available indoors for self-service compared to the 36% of centers observed by Benjamin Neelon et al.²⁵

Center directors reported that staff sit with children and consume similar foods more frequently than what was observed in North Carolina²⁵ and what has been reported by directors from western states.^27,28 Family-style service, in which children serve themselves, was reported in 38% of centers, which is consistent with a report by Sigman-Grant et al. in four western states,²⁷ but more than what has been observed in North Carolina.²⁵ Parent-provided lunchbox service was reported in only 11% of centers. Whereas 83% of center directors indicated that staff followed best practice of gently encouraging children to try new or less favorite foods, only 1% of centers indicated that staff followed best practice of helping children determine if they are still hungry before serving seconds. Annual nutrition training of staff (31% of centers) was lower than what has been reported in North Carolina²⁵ (48%), and similar to what has been reported in states from the West.²⁸

This study has some limitations. Direct observation is the gold standard for evaluating nutritional practices and comparing foods and beverages served at meals versus snacks over the course of a week, but we were not able to conduct in-person assessments in this large sample of centers. Numerous published studies^27–29,38 have used surveys to elicit nutrition practices served in child-care settings, although surveys may be subject to self-report bias and nonresponse bias. Our data suggested, in fact, that directors may have underreported the frequency of service of foods with added sugars and sodium. On the survey, 16% of center directors reported serving salty foods such as crackers, and 12% reported serving sugary foods such as cookies, at least three times a week at snack, but menu analysis indicated that these foods were served at least three times weekly at snack in 87% of centers. Similarly, on the survey, 169 (65%) of centers reported serving vegetables other than corn, green beans, or potatoes at least once a day compared to 6 (4%) of centers according to the lunch and afternoon snack menus (data not shown). Centers that elected not to provide menus may have differed in foods or beverages served from those centers that did provide menus; however, a demographic analysis of centers providing menus compared to those that did not showed no significant differences with the exception of being affiliated with a national chain. One study³⁶ in a single child-care center has suggested that menus are an unreliable data source of the exact food or beverage actually served. However, a larger study²⁶ of 84 centers found menus to be a reasonably reliable (87% agreement) indicator of foods actually served, particularly for food types. This study did not measure child consumption of foods and beverages—merely provision. Children's consumption may differ from what is actually provided to them, although one previous study suggests that children consume most of the foods and beverages served in child care.¹⁵ Future studies may wish to explore foods and beverages consumed by children in care and assess the relationship of consumption to weight status. Last, while the nutrition practices and types of foods served over the course of the day were consistent with what has been reported in child-care settings in other regions of the country, our findings with regard to the differences in meals compared to snacks need to be replicated in other regions. It is also possible that secular changes in local or school-based catering services may have improved the nutritional quality of foods and beverages served since this study was conducted, although we are unaware of any initiatives that have addressed the offerings at meals versus snacks, as this difference has not previously been reported. It is important to note that we did not assess barriers and facilitators to serving healthier foods and beverages, nor did we ask why centers served specific foods rather than others.

Conclusion

Because most children are in child care and consume many of their daily meals and snacks in these environments, these settings offer a venue to prevent childhood obesity. Our data suggest that despite prevention efforts to improve children's diets in child care, meals—and in particular snacks—still do not provide adequate servings of whole fruits and nonstarchy vegetables, and are comprised of foods and beverages with added sugars, fats, and sodium. Theoretically, there is no reason for the nutritional components of snacks to differ from lunches. While snacks are generally smaller than meals, they are an integral part of preschool-aged children's diets, and constitute, on average, 26% of children's daily energy intake.³⁹ Thus, it would seem reasonable to expect that children would be consuming a similar proportion of their weekly intakes of fruits, vegetables, and meat/meat alternatives at snack times; yet these data suggest this is not happening for children in full-day child-care centers. The USDA is currently revising its guidelines for the Child and Adult Care Food Program (CACFP) with regard to the nutritional components of meals and snacks in child care as a part of the Healthy, Hunger-Free Kids Act (2010). The revised guidelines are expected to align closely with an Institute of Medicine report³⁹ released in the fall of 2010 that calls for snacks served in participating centers to consist of two food group components—milk, fruit, nonstarchy vegetable, grain, and meat/meat alternative—and for each of these food groups to be served at snack at least twice per week. These data suggest these requirements will present a sharp departure from the snacks currently served in child-care centers, which typically consist of 100% juice and a refined grain.

Children of this age are establishing taste preferences and dietary habits,^40–42 and developing a cultural understanding of acceptable behaviors and the types of foods that are consumed at meals and snacks.^43,44 Given the extensive body of research about the importance of repeated exposures to new foods for establishing healthy eating habits, it seems plausible to infer that there is a similar need to expose young children to different types of food groups at snacks (including fruits, vegetables, and meat/meat alternatives) to establish healthy snacking habits. Dietitians and menu planners may consider the messages young children in child care may be inadvertently receiving regarding the types of foods provided at meals versus snacks, and how these children may be prone to eat more at snack times, when highly palatable foods are often served, and less at meals when fruits and vegetables are served. Regardless, these data suggest that snacks represent a possible missed opportunity for consumption of essential micronutrients and macronutrients and deserve further attention in menu planning and subsequent research.

Acknowledgments

This research was supported in part by a grant under the Robert Wood Johnson Foundation Physician Faculty Scholars Program, a career development award from the National Institutes of Health (K23 HL088053), and Rapid Response Award from the Robert Wood Johnson Foundation's Active Living and Healthy Eating Research programs. We are also grateful to 4C for Children, the child-care resource and referral agency for Greater Cincinnati and Dayton, for their assistance with this project. Last, we are grateful to Robert Tamer, MPH, MPA, who assisted with data analysis, Hadley Sauers who assisted with data collection, and Sabine Johns, who planted the seed for this research endeavor.

This study was supported by a National Institutes of Health grant, K23 HL088053, and Robert Wood Johnson Foundation Faculty Scholars Award to Dr. Copeland, and a Robert Wood Johnson Foundation Active Living and Healthy Eating Research Rapid Response Award to Dr. Wosje.

Author Disclosure Statement

No competing financial interests exist.