Consistency of Compliance with Nutrition-Related Regulations among Delaware Child Care Centers

Abstract

Background:

As more calls are made in the literature for nutrition interventions to be delivered in child care settings, research on the implementation of these interventions becomes more important. This study examined compliance with Delaware's regulations related to nutrition in child care settings, which are designed to improve the nutrition-related environment in these settings.

Methods:

A stratified random sample of licensed child care centers (n=233) was created from the total population of eligible centers in Delaware (N=450). Study staff visited each center and distributed self-administered surveys to the director and two randomly selected teachers. Surveys contained items about classroom-level compliance with the regulations along with center-level characteristics. Bivariate analyses were conducted to explore relationships between consistent compliance with each regulation component and center-level characteristics.

Results:

A total of 179 of the 233 centers in the selected sample participated in the study. Compliance with the regulations varied within centers and across components; the highest levels of consistent compliance were reported for juice type (88.3%) and the lowest levels for whole grains (18.6%). Center characteristics, such as participation in the Child and Adult Care Food Program, were associated with consistent compliance for certain regulations components.

Conclusions:

Our results suggest that these types of regulations can be implemented across a diversity of centers, but that certain components (e.g., those relating to whole grains and water) may need further clarification. Our results also suggest that there are certain types of centers on which to focus training efforts to maximize compliance.

Introduction

As childhood obesity prevention efforts gain in number and reach, evolving from a focus on individual-level interventions into more setting-based, community- and policy-level efforts, research into the translation and implementation of these interventions becomes more important.¹ Numerous calls have been made in the literature for more nutrition or physical activity interventions, or both, to be delivered in the child care setting.^2–6 These reports have identified the increasing rates of childhood overweight and obesity along with the realization that many young children spend a large number of hours each week in nonparental care.

Delaware is one state that has capitalized on the opportunity to promote healthy development in child care settings, specifically focusing on policy-level interventions that have the potential for population-wide effect. In 2007, the Delaware Office of Child Care Licensing (OCCL) increased the nutrition- and physical activity-related standards for licensed child care providers.^7,8 The new regulations (known as the Delacare Rules) focus on providing high-quality, nutritious foods to children in these settings and providing ample opportunity for children to engage in physical activity.

In the time since the regulations were put in place, several entities have provided training and technical assistance to providers in an attempt to encourage and increase compliance. Nemours Health and Prevention Services (NHPS), the population health arm of the Nemours pediatric health care system located in the Delaware Valley, is one entity that has been involved in these efforts statewide.⁸ To monitor progress toward 100% compliance with the regulations, evaluate efforts to increase policies and practices consistent with the regulations, and identify additional training needs of providers, NHPS has administered a statewide survey of child care providers biennially since 2006. The most recent iteration of this survey was administered to a statewide, representative sample of providers in Delaware in the fall of 2011.

The current study uses these data to explore the extent to which practices within child care centers serving young children (predominately aged 3 and 4 years) are compliant with Delaware's regulations related to nutrition and whether compliance is consistent within the center using self-report of practices from two randomly selected classroom teachers within each center. Our aim was to provide information about the characteristics that were associated with consistent compliance within centers to aid in both understanding implementation of the regulations and the ways in which additional training and technical assistance can be tailored to the centers that need it most.

Methods

Study Design

Stratified random sampling was used to generate a representative sample of licensed child care centers in the state of Delaware. First, a complete listing of all licensed child care centers (full or part time) in operation in August, 2011, was obtained from the Delaware OCCL (N=450). To include the appropriate distribution of centers from across the state, we randomly selected 235 centers from a list stratified by geographic location (City of Wilmington and the three Delaware counties—Kent, Sussex, and New Castle County—outside the city of Wilmington). All Head Start centers were included in the sample. All selected centers (n=235) were contacted by phone and invited to participate.

A trained survey administrator then visited each site that agreed to participate during the fall of 2011 to administer questionnaires to the director and two randomly selected teachers. Center directors were eligible for inclusion if they were the center director or head administrator of a selected licensed child care center. Teachers were eligible if they were currently working with toddlers, preschoolers, and/or school-age children in the same centers. Because teachers spend the most time with the children at child care settings, specifically during meal times, they were believed to be the most knowledgeable staff to answer questions regarding eating habits. Our team considered using food service personnel as the key informant for food-related items, but not all centers have food service staff (only 67% in our sample, for example), and we wanted to ensure consistency of data quality across centers. Teachers who participated in the study were given a $10 gift card as an incentive.

Questionnaires were printed on scannable forms, scanned into a database, and reviewed by the study team for inconsistencies.

Study Measures

The questionnaires were designed to collect information from center-based providers on nutrition- and physical activity–related knowledge, policies, and practices. Both questionnaires were reviewed by content experts to ensure face validity and pilot tested with three centers prior to using with the entire sample.

Director questionnaire

The director's questionnaire measured awareness of current nutrition and physical activity standards, staff behaviors, and staff training at the center level. The director version included multiple-choice questions on meal preparation and planning and the center's guidelines and policies related to nutrition, staff training, and behavior.

Teacher questionnaire

The teacher's questionnaire contained multiple-choice questions about classroom characteristics, healthy eating and physical activity practices, and behaviors within their classrooms.

For our analysis, each classroom was considered to be in compliance with the policy if the teacher reported practices that were equal to or better than the policy for a given domain. For example, one component of the nutrition regulations allows sweet grains to be served to children once in a 2-week cycle. If a teacher reported that sweet grains were never served, that classroom would be considered in compliance (along with teachers that reported practices exactly in line with the policy, in this case, once in a 2-week cycle). Additional variables were created for each domain representing whether or not both teachers reported consistent compliance. Our research design considered the independent report of two teachers to be a proxy for center-wide (consistent) compliance with the policy. We hypothesized that if the two randomly selected teachers' responses were the same, those centers were more likely to have consistent compliance compared with centers where the two teachers selected dissimilar answer choices (i.e., inconsistent compliance). Inconsistent compliance included both centers where the teachers reported differently (e.g., one reported being in compliance and the other reported not being in compliance) and centers where both teachers reported being noncompliant. We considered alternate analysis plans (e.g., comparing consistently noncompliant with others), but ultimately hypothesized the greatest opportunity for impact of these policies on children would be seen in consistently compliant centers and that consistent compliance represents full implementation of these interventions at the center level. We sought to examine the predictors of this full implementation among participating centers.

Statistical Analysis

Given the sampling plan used for the study, the data were weighted to compensate for the differential probability of selection and the differential response rates within each of the four geographic areas. The unit of analysis was the centers (unweighted n=176). All analyses were conducted using the Complex Sampling module of SPSS (IBM, Armonk, NY) version 20.0 for Macintosh.

First, descriptive analyses were conducted to describe the study population and the predictor and outcome variables. Second, a series of cross tabulations was run to examine associations between consistent compliance with each regulation component and various center-level characteristics. For these calculations, the dependent variables were binary (0,1) variables indicating consistent compliance for a particular component reported by the randomly selected teachers within each center (e.g., 1=centers where both teachers reported compliance; 0=centers where teachers reported inconsistent compliance or consistent noncompliance). Independent variables considered included center type (e.g., independent, franchise), whether there was a food service staff person on site (yes or no), the location of food preparation (on site, central kitchen, other), and the person(s) responsible for menu planning (director/owner or other). The independent variables were reported by the center director. A chi-squared test for independence of rows and columns for weighted data was calculated for each binary association; we report the p value for these tests along with odds ratios (OR) for associations with p≤0.05.

This study was reviewed by the Nemours Institutional Review Board and was given exempt status. Study procedures complied with requirements related to participants' privacy and confidentiality; written documentation of informed consent was obtained from all participants.

Results

Ultimately, 179 centers out of the 233 in the database participated in the survey (76.17% response rate). At three of the centers, there were incomplete data from at least one of the two teachers, thus they were excluded from the sample, and the final sample size was 176 centers. Nearly half of the participating centers were independently (privately) owned and operated (49.6%), and 57.4% reported participating in the Child and Adult Care Food Program (CACFP), a federal nutrition education and meal reimbursement program for child care settings (Table 1). The majority of centers (78.9%) reported that the food served in the center was prepared on site. In terms of food preparation, while 67.1% of centers reported having one or more food service staff members on site, only 25.6% reported that these employees were responsible for menu planning.

Table 1.

Characteristics of Participating Centers^a

	Centers n (%)^b
Type of center (n=165)^c
Head Start or Early Childhood Assistance Program (Early Head Start)	26 (16.0)
Nonprofit	53 (32.0)
Franchise	15 (8.5)
Independently owned and operated	80 (49.6)
CACFP participant	95 (57.4)
Food service staff on site (n=162)	107 (67.1)
Where food is prepared (n=155)
Prepared on site	120 (78.9)
Prepared by a central kitchen	17 (10.1)
Other	18 (11.0)
Responsibility for menu planning (n=170)^c
Director/owner	96 (57.1)
Food service staff	43 (25.6)
Other (e.g., nutritionist, corporate headquarters, teacher)	59 (35.0)

Total n may not add up to 176 because of missing data for a particular item.

Unweighted n and weighted percentage.

Respondents were allowed to choose more than one answer option.

CACFP, Child and Adult Care Food Program.

Compliance with the regulations varied significantly across components and within centers (Table 2). The highest levels of consistent compliance were seen for juice type (88.3%) and water availability inside the center (82.1%). The lowest levels of consistent compliance were seen for whole grains (18.6%), water availability outside the center (35.6%), and fried or prefried meats (51.4%).

Table 2.

Consistency of Teacher-Reported Compliance by Center (n=176 Centers^a)

	Classroom 1 compliant n (%)^b	Classroom 2 compliant n (%)^b	Consistent compliance n (%)^b
1. Fried or prefried and then baked vegetables (n=157)	120 (76.9)	119 (75.2)	103 (65.0)
2. Processed meats (n=163)	135 (83.1)	137 (84.4)	119 (73.3)
3. Fried or prefried meats (n=163)	104 (64.4)	105 (65.0)	83 (51.4)
4. Cheese type (n=162)	136 (83.5)	129 (80.9)	114 (70.8)
5. Whole grains (n=164)	58 (35.3)	55 (33.4)	30 (18.6)
6. Sweet grains/breads (n=159)	127 (79.8)	136 (86.1)	117 (73.7)
7. Juice type (n=160)	150 (93.9)	147 (92.6)	140 (88.3)
8. Juice quantity (n=161)	135 (82.8)	140 (85.8)	121 (73.5)
9. Water availability inside (n=167)	152 (90.2)	149 (89.4)	138 (82.1)
10. Water availability outside (n=162)	89 (54.8)	86 (52.9)	58 (35.6)

Centers were included in each row only if they had complete data for that component.

Unweighted n and weighted percentage.

We examined characteristics that were associated with consistent compliance with each regulation component via a series of cross tabulations and chi-squared tests (Table 3). Our results suggest that Head Start programs were more likely than others to report consistent compliance with the fried or prefried vegetable (OR 3.0, p=0.035) and sweet grain regulations (OR 6.8, p=0.005). Franchise centers were significantly more likely to report consistent compliance with the fried or prefried vegetable regulation as well (OR 4.98, p=0.026). In general, independently owned and operated centers were less likely to comply with the regulations, except for the regulation related to water availability outside, where they were nearly twice as likely to report consistent compliance compared with other centers (p=0.048). Centers that participate in CACFP were less likely to report consistent compliance with the processed meats regulation (OR 0.47, p=0.04) but more likely to report consistent compliance with the whole grains regulation (OR 3.9, p=0.006). The location of food preparation (e.g., on site, other) was only associated with consistent compliance with the whole grain component (OR 4.4, p=0.04) for centers with on-site food preparation compared with others. Centers where food service staff were responsible for menu planning were significantly less likely to report consistent compliance with the fried or prefried vegetable component (OR 0.36, p=0.006) and fried or prefried meat component (OR 0.31, p=0.003). These associations were similar to those found related to the presence of food service staff on site in general.

Table 3.

Predictors of Consistent Compliance with Delacare Rules

	Regulation component^a OR (p value for test of independence of rows and columns^b)
	1	2	3	4	5	6	7	8	9	10
Type of center
Head Start or Early Childhood Assistance Program (Early Head Start)	3.0 (0.035)	(0.51)	(0.16)	(0.23)	(0.55)	6.8 (0.005)	(0.10)	(0.20)	(0.06)	(0.055)
Nonprofit	(0.82)	(0.13)	(0.62)	(0.57)	(0.79)	(0.70)	(0.019)	3.15 (0.006)	(0.62)	(0.08)
Franchise	4.98 (0.026)	(0.80)	(0.23)	(0.27)	(0.07)	(0.73)	0.27 (0.017)	(0.89)	(0.96)	(0.58)
Independently owned and operated	0.30 (<0.001)	0.43 (0.016)	(0.06)	(0.34)	(0.56)	0.32 (0.002)	(0.16)	0.39 (0.009)	(0.88)	1.96 (0.048)
CACFP participant	(0.26)	0.47 (0.04)	(0.51)	(0.35)	3.9 (0.006)	(0.08)	(0.11)	(0.97)	(0.90)	(0.64)
Food service staff on site	0.31 (0.003)	(0.10)	0.30 (0.001)	(0.88)	(0.10)	(0.16)	(0.75)	(0.52)	(0.97)	—
Where food is prepared
Prepared on site	(0.73)	(0.71)	(0.91)	(0.74)	4.4 (0.04)	(0.83)	(0.56)	(0.19)	(0.48)	—
Other^c
Who is responsible for menu planning
Director/owner^c
Food service staff	0.36 (0.006)	(0.11)	0.31 (0.003)	(0.12)	(0.86)	(0.29)	(0.14)	(0.26)	(0.33)	—
Other (e.g., nutritionist, corporate headquarters, teacher)^c

Numbers correspond to regulation component numbering scheme from Table 2.

Bold text indicates p≤0.05.

Comparison group.

Discussion

Consistent compliance, within classrooms (e.g., from day-to-day) and within and across facilities, is the goal for these and other regulations passed at the state level. Our study suggests that while compliance with the majority of the regulation components was relatively high as reported by individual teachers, consistent compliance across both classrooms was lower. The difference between the percentage of centers that would have been classified as compliant based on one teacher report and those that would be classified as compliant without consistent report of compliance varied from 5.6% (93.9% vs. 88.3%) for juice type to 16.7% (35.3% vs. 18.6%) for whole grains. In every case, using percent agreement as the definition of compliance for the center represented at least a 4.3% reduction in the proportion of centers reporting compliance. This is relevant to those seeking to understand and increase compliance with these policies as well as those studying the implementation of these policies. Our results suggest that using a single teacher's report as a proxy for center level compliance with nutrition policies will result in a higher estimate of compliance compared with using multiple reports, and that the magnitude of this difference varies across different types of nutrition rules. Furthermore, our finding of low compliance across classrooms indicates that child care centers should consider implementing further training, monitoring, and technical support to help ensure consistency of compliance at the center level.

When we examined the characteristics related to center-level (consistent) compliance we found that the whole grain component and the outdoor water availability component represented two of the largest challenges to centers. Interestingly, we found that consistent compliance with the whole grain component was considerably lower than any other component despite the fact that compliance with this component requires the same process as compliance with other nutrition-related items, namely, menu planning processes that include the healthier option and purchasing those healthier items. Previous research suggests that certain nutrition-related policies are less understood by providers,⁹ that cost or availability might be an issue,¹⁰ or both. Other potential explanations worth considering are that purchasing and serving whole grains daily is a challenge for providers and that providers or children may be less inclined to purchase or consume whole grain products. The CACFP participants and those with on-site food preparation were significantly more likely to report consistent compliance with this component. Future training efforts should focus on improving understanding and implementation of this component among non-CACFP centers, perhaps involving practice reading ingredient lists to identify whole grains, taste testing whole grain products, or working through vendors to increase availability of whole grain products.

We also found lower (35.6%) consistent reported compliance with water availability outside, which is consistent with the limited existing research on water availability in child care settings.¹¹ Compliance with this component requires more structural changes (e.g., installation of water fountains) than the other components, and it is possible that centers have been slower to make these types of changes. We found that centers that are independently owned and operated were more likely to comply with this regulation component (OR 1.96). It is possible that independent centers can more easily make the physical changes to their facilities required for compliance with this component because decisions about changes to physical spaces are made at the center level (as opposed to at the corporate level). Additional research should be done to determine the barriers to compliance with the policy and to determine the policy and programmatic ways to increase availability of water in outdoor play areas.

In general, centers that were independently owned and operated were less likely to report consistent compliance with the regulations compared with other centers (e.g., those that were members of franchises or groups of centers). This may be due to the fact that these centers have less access to resources and materials necessary to understand and implement the regulations. Again, future training and technical assistance work may want to prioritize these centers if resources are limited. A network of independently owned and operated centers may contribute support similar to the resources and peer support provided by the franchise centers.

Last, our analysis suggests that centers with food service staff on site and those where food service providers are responsible for meal planning were less likely to be in compliance with fried food regulation components. We hypothesize that centers with food service providers may be larger or have cooking facilities, which allow for preparation of fried foods (either through baking prefried foods or frying them at the center), and that the centers without these staffs (and, therefore, these facilities) may be serving cold, prepared items that are less likely to be fried/prefried (compliance with this regulation included that fried/prefried foods never be served, which is more stringent than the regulations). Future research may want to specifically investigate the influence of food service staff and cooking facilities on centers' willingness and ability to comply with nutrition regulations. For example, research concerning food service personnel's attitudes toward these types of regulations may help identify the ways in which they influence compliance.

Limitations

This study had a number of limitations that may influence interpretation of results. First, while we had a very high response rate (76%) and are confident about the applicability of our results to the population of providers in Delaware, the generalizability of our findings to centers outside of Delaware is unknown. Second, while we feel that the addition of a second teacher report increases the validity of our compliance measures, self-report is prone to social desirability bias in ways that may have inflated our results. If, for example, teachers from franchise centers were more likely to know about the regulations and report compliance (regardless of actual practice in the classroom), our results may suggest a stronger relationship between franchise centers and compliance than actually exists. Future research may want to also include an observational assessment, if feasible, to improve validity. Despite these limitations, we feel our study contributes significantly to the literature on the implementation of these types of interventions at the center level.

Conclusions

Our results suggest a number of starting points for policymakers interested in implementing similar regulations. First, we found that reported compliance was relatively high for the majority of regulations, suggesting that these regulations can be implemented across a diversity of centers. Second, certain components, such as those relating to whole grains and access to water outside, may need either further clarification by policy makers or additional training and technical assistance efforts, or both. Our results suggest a number of possible child care center characteristics on which to focus these efforts, such as emphasizing efforts within independently owned and operated centers or within non-CACFP centers.

Given that we found a significant difference between individual teachers' report of compliance within their classroom and the level of consistent compliance within centers, other researchers attempting to classify compliance with these and similar policies may want to consider using multiple data points for each center. Relying on self-report from one teacher may overestimate compliance at the center level and mask potential challenges related to compliance or differences between compliance and other variables (e.g., training).

Footnotes

Acknowledgments

Funding for this study was provided by the Robert Wood Johnson Foundation. The authors wish to thank Roberta Gealt and Katherine Ware from the Center for Drug and Alcohol Studies at the University of Delaware for coordinating recruitment and data collection for this study and the hundreds of child care providers in Delaware that participated in the study. The authors would also like to thank Theresa Michel for editorial support.

Author Disclosure Statement

No competing financial interests exist.

References

Erinosho

, Hales

, McWilliams

et al. Nutrition policies at child-care centers and impact on role modeling of healthy eating behaviors of caregivers. J Acad Nutr Diet, 2012; 112:119–124.

Benjamin

, Cradock

, Walker

et al. Obesity prevention in child care: A review of US state regulations. BMC Public Health, 2008; 8:188.

Koplan

, Liverman

, Kraak

. Institute of Medicine. Preventing Childhood Obesity: Health in the Balance. The National Academies Press: Washington, DC, 2004.

Institute of Medicine. Local Government Actions to Prevent Childhood Obesity. Institute of Medicine: Washington, DC, 2009.

Khan

, Sobush

, Keener

et al. Centers for Disease Control and Prevention. Recommended community strategies and measurements to prevent obesity in the United States. MMWR Recomm Rep, 2009; 58:1–26.

Kreichauf

, Wildgruber

, Krombholz

et al. Toy Box-study group. Critical narrative review to identify educational strategies promoting physical activity in preschool. Obes Rev, 2012; 13,Suppl 1:96–105.

Delacare Rules for Early Care. Education, School-Age Centers, January 1, 2007. Delaware Code, Title 31, Chapter 3, Subchapter III, Subsections 341–345.

Van Stan

, Lessard

, Dupont Phillips

. The impact of a statewide training to increase child care providers' knowledge of nutrition and physical activity rules in Delaware. Child Obes, 2013; 9:43–50.

Lanigan

. The relationship between practices and child care providers' beliefs related to child feeding and obesity prevention. J Nutr Educ Behav, 2012; 44:521–529.

10.

Monsivais

, Johnson

. Improving nutrition in home child care: Are food costs a barrier? Public Health Nutr, 2012; 15:370–376.

11.

Patel

, Hampton

. Encouraging consumption of water in school and child care settings: Access, challenges, and strategies for improvement. Am J Public Health, 2011; 101:1370–1379.