Influential factors of caregiver behavior at mealtime:
A study of 24 child-care programs


Author: Nahikian-Nelms, Marcia
Source: American Dietetic Association. Journal v97n5 (May 1997): 505-509
ISSN: 0002-8223
Number: 03265264
Copyright: Copyright American Dietetic Association 1997

Objective To measure the nutrition knowledge and attitudes of caregivers in child-care programs and to observe the behaviors of caregivers as they interact with children at mealtime.

Design A nonexperimental research design. Nutrition knowledge and attitudes toward nutrition were measured using survey methodology designed for this study. Observers measured caregiver behavior by means of observation and quantified it using a behavior checklist. Interrater reliability was 98%. Instruments were piloted and content validity was established. Using Cronbach's of, reliability was .69 and .67, respectively.

Subjects/setting Participants were 113 caregivers in 24 licensed child-care programs in three counties in Illinois. Statistical analyses performed The independent variables (caregiver nutrition knowledge, caregiver nutrition attitudes, years of teaching, prior nutrition training, and education level) were correlated to the dependent variable (caregiver behavior), using the Pearson product moment correlation. The combination of variance from all independent variables was analyzed using the general linear regression model. Results Although caregivers held beliefs that should have a positive influence on children's eating behaviors, they demonstrated low knowledge of nutrition and displayed behaviors at mealtimes that were inconsistent with their beliefs and expert recommendations. Positive correlations were found between nutrition knowledge and behavior at mealtime, nutrition knowledge and attitudes, and attitudes and caregiver behavior.

Applications These instruments can be used to assess and teach nutrition practices through self-assessment, training, and coursework. Results also begin to define the role of the caregiver in children's nutrition practices. Results demonstrate that caregiver behavior can be enhanced by addressing nutrition knowledge and attitudes. JAm Diet Assoc. 1997;97:505-509.

The growth and development of children are not only influenced by proper early nutrition but also by the establishment of healthful food preferences and eating behaviors. The development of lifelong eating behaviors is shaped through a multifactorial process (1-10) that includes innate preferences, psychosocial and cognitive development, familiarity, and the social-affective context provided by peers, teachers, and parents. Early childhood caregivers have become new, key players in this dynamic.

Much of the recent research assessing nutrition practices in early childhood programs has focused on the adequacy of foods that are served (11,12). A holistic approach to nutrition practices, however, addresses not only the actual foods that are served but also the physical environment, caregiver behavior during meals, and the integration of developmentally appropriate nutrition education into the early childhood curriculum. This study sought to determine whether attitudes and behaviors that have been identified as crucial in shaping the development of food preferences and eating behaviors are typically practiced in child-care programs (1-6). This study also attempted to measure the presence of appropriate caregiver behaviors at mealtime and describe the relationship between caregiver attitudes toward nutrition, nutrition knowledge, and caregiver behavior.

METHODS

Instrumentation

This study involved the development and use of three instruments.' The nutrition knowledge instrument was adapted with permission from a 20-item, multiple choice test, adapted by Drake (11). The nutrition concepts for this instrument were chosen because of their role in maintaining the health of young children. Questions addressed common nutrition problems and the tools used to plan a healthful diet, including the Recommended Dietary Allowances, the US Department of Agriculture (USDA) Food Guide Pyramid, and the USDA Child and Adult Care Food Program guidelines (13-15).

(Table Omitted)

Captioned as: Table

The second instrument was an attitudinal inventory containing 27 items chosen to address the role of nutrition in the childcare program, the role of the caregiver in nutrition practices and their beliefs about early childhood nutrition, and mealtime rules that might influence the development of food preferences and eating behaviors.

Each instrument was pilot tested twice and revised. A panel of experts, which included two registered dietitians and four faculty members in child development and early childhood education, established content validity of both instruments by responding to items regarding their own area of expertise. Internal reliability was established using Cronbach's a at .67 and .69, respectively.

The final instrument was a behavior checklist for use during the observation of meals. The behaviors selected for observation were chosen on the basis of their being important areas of concern regarding interaction between caregivers and children during meals (3,12,16). These behaviors have also been incorporated into the child-care standards outlined by The American Dietetic Association (ADA), the Society for Nutrition Education (SNE), the National Association for the Education of Young Children (NAEYC), and the Headstart Program (17-20). Interrater reliability was established at 100% before the onset of the study and 96% during the study.

Subjects

Directors for all licensed child-care programs (n=27) in three counties in rural southern Illinois were contacted, informed about the purpose of the study, and offered a free workshop for all participants at the conclusion of the study. Three programs declined to participate. The 113 caregivers who agreed to participate (100% acceptance) ranged in age from 18 to 66 years of age (mode=22 years). The mean amount of teaching experience was 5.8 years.

Protocol

Before the observations began, a training session was held for the four nutrition graduate students who assisted in data collection. The training consisted of review of instruments, role playing of data collection, and group problem-solving. On the day of the visit, each staff member who had direct childcare responsibility at mealtime was given a cover letter explaining the study and its purpose. Each participant was asked to write a self-selected code on a sticker name tag, which they wore during the meal observation. If a duplicate code was chosen, the participant was asked to select another code. One of four trained observers monitored the luncheon meal and tallied caregiver behaviors using the behavior checklist previously described. Several trips were made to each center so that all classrooms could be observed. Each caregiver was given the knowledge and attitude instruments to complete independently after the meal observation. The caregiver's code was used to connect observation data with survey data.

Statistical Methods

This study resulted in four sets of data: nutrition knowledge scores, nutrition attitude scores, demographic characteristics, and caregiver behavior scores. The first three independent variables were correlated with the fourth dependent variable (caregiver behavior score) using the Pearson product moment correlation. Additionally, the demographics were analyzed using descriptive statistics. Demographic characteristics are summarized in Table 1.

The combination of variance from all independent variables was analyzed using the general linear regression model. An F statistic based on the regression equation was calculated to detern ine significance at .05 alpha level.

RESULTS

Generally, the 113 caregivers in this study seemed to agree with the behaviors and attitudes that have been identified as important in fostering good eating habits. Overall, the teachers scored 69 out of a possible 81 points on the attitudinal inventory. A high score on this instrument meant that the caregiver held beliefs that should have a positive influence on the children's development of food preferences and appropriate eating behaviors. Yet despite evidence of these attitudes, the caregivers studied showed an overall low knowledge of nutrition and displayed behaviors at mealtimes that were inconsistent with their attitudes and beliefs about nutrition and inconsistent with expert recommendations.

Although 69% of the sample listed college coursework as their source for nutrition training, the group's overall nutrition knowledge remained low. Participants registered a mean score of 10.9 out of a possible 20 points (see Table 2) on the nutrition knowledge instrument. Less than 50% of respondents answered 10 of the questions correctly. The responses to items on the nutrition knowledge instrument point to several misconceptions regarding the science and application of nutrition. Only 13% correctly defined the Recommended Dietary Allowances or understood the correct application of its reference values. Only 19% correctly related the intake of sugar to tooth decay, and 80% thought sugar caused diabetes mellitus. Many were unable to identify the best sources of vitamin A, protein, and iron. Forty-nine percent of the sample did not know that two thirds of the nutrition needs of a child should be met by the child-care program when the child attends for 8 hours or more per day.

There was a positive correlation (r=.18, P=.04) between caregivers' nutrition knowledge and caregiver behavior at mealtime. This may indicate that those caregivers with the highest score on the nutrition knowledge instrument were the most likely to have the highest behavior scores. Years of teaching experience and caregiver behavior were correlated (r=.19, P=.03) positively as well. This may be due to the fact that an experienced teacher may be better able to integrate nutrition into all parts of the curriculum, which would optimally include meal and snack times. There was a statistically significant relationship (r=.27, P=.003) between level of education and nutrition knowledge, which suggests that those persons with higher levels of education also scored higher on the nutrition knowledge instrument. There did not seem to be a relationship between previous nutrition training and the caregiver's attitudes toward nutrition (r=.02, P=.81). Previous nutrition training and level of education were also not related to caregiver behavior (r=.12, P=.17; r=.15, P=.11). This may reflect the type of nutrition training that is offered to early childhood educators.

The concept of role modeling was an important part of the attitudinal inventory and the behavior checklist. The interpretation of role modeling, which can have a notable influence on the specific foods children choose to consume, has guided the recommendations that caregivers should sit with children at meals and consume the same food as the children (1,2,17-21). Approximately 86% of the respondents agreed with the statement, "When I eat with children, I influence their eating habits." The respondents also overwhelmingly agreed (95%) with the statement: "Caregivers should sit with children during meals in order to model good eating habits."

However, only 75% of the respondents agreed with the statement: "It is important for caregivers to consume the same food as the children during meals and snacks." These behaviors were also noted during the observations of caregivers during meals. The caregiver did sit down with the children at meal time in 69% of the observations. But in only 53% of these observations where the caregiver sat with the children did the caregiver consume the same food as the children. Caregivers who did not consume the same food as the children often did not eat at all, and if they did it was often fast food and sodas. Many caregivers were observed to drink sodas during the meal even if they consumed the same food as the children did.

(Table Omitted)

Captioned as: Table

(Table Omitted)

The social environment in which children eat can also have a dramatic effect on eating behaviors. If mealtime is used as a time of conflict or discipline, children may associate eating with those same feelings. It was noted that caregivers engaged in conversation with children only 74% of the time. That meant that in some instances even though the caregiver was sitting with the children no conversation occurred. In some observations the only interaction between caregiver and child involved directives such as: "You need to eat."

Caregivers may not find it easy to integrate nutrition into the curriculum. Although 83% of the respondents agreed that mealtime should be used as an opportunity to educate children about nutrition and 86% of the respondents agreed that mealtime could be integrated into their teaching curriculum, only 50% of those caregivers observed made any comment about nutrition during the meals.

Approximately 90% of the caregivers believed children will learn to eat a variety of foods if they are offered to them. This is consistent with the literature and supported by the recommendations of ADA, SNE, and NAEYC (17-20). Much of the research by Birch et al has focused on the role of exposure on the development of food preferences (4-6). Birch's work has determined that approximately 10 to 15 exposures of a novel food will enhance the preference for that food. Because one of the basic principles of ensuring good nutritional habits is based on consuming a variety of foods, the first step in ensuring a child's intake is to offer that variety. Parents and caregivers often allow the child's initial rejection of a novel food to determine whether that food is offered again ( 16). Thirty-eight percent of the caregivers agreed with the statement that it is best to substitute a food the child likes better if the child is refusing to eat. But research has shown continued exposure to a food will increase the likelihood that the child will eventually consume it. (4-6)

For several items, respondents' answers were not consistent with the reconuuended response. Two items addressed the children's role in serving themselves and in assisting with table setting and clean up. Only 3 of the 24 centers observed used a family-style foodservice. The other 21 programs served the children plates of food that were prepared ahead of time. Recommendations from ADA, SNE, and NAEYC all suggest that family-style foodservice is preferable (17-20). This type of foodservice allows children to learn the coordination of physically manipulating utensils. It also supports the belief that children have a strong innate control for their energy requirements (11). It is recommended that caregivers allow children to realize when they are hungry and when they are full. It is the caregiver's responsibility to provide nutritious food choices but the child should determine the amounts eaten (16). Family-style foodservice supports these beliefs.

Two items addressed whether there should be mealtime rules designed to encourage children to taste all foods. There seemed to be disparity among these responses. When asked about the statement, "Children should be required to taste all the foods served," only 47% of the caregivers agreed. Birch found that preference did not increase for a novel food unless it was tasted (6). This would seem to support the belief that we should encourage tasting of all foods. Mealtime observations determined that only 59% of the caregivers observed asked the children to taste all the foods offered.

The use of contingencies to manipulate a child's food intake has been the focus of much research. Birch et al (5) determined that children actually decreased their preference for a food that was used as the contingency for the reward. In a study by Stanek et al (21) 56% of the parents interviewed said that they used candy and dessert foods to reward their children for eating a meal. In the current study, only 14.2% of the caregivers agreed with the statement: "If a child refuses to try a particular food, it will help to encourage tasting the food by promising dessert or another treat." This is consistent with the actual observations where only 7% of the caregivers used food as a reward, punishment, or pacifier.

Caregiver behaviors, as shown in the Figure, were fairly consistent with the optimal recommendations put forth by ADA, SNE, and NAEYC (17-20). Observations indicated that caregivers achieved a mean score of 6.59 out of a possible 10 points. The largest variances included the absence of familystyle meals, the lack of interaction between caregivers and children at meal times, and the lack of nutrition education.

Caregiver behavior and attitudes toward nutrition were positively correlated (r=.23, P=.01). It appears that if the caregiver agreed with these beliefs they interacted with children at mealtime in a way that would provide for the optimal environment. Additionally, nutrition knowledge and attitudes were statistically correlated (r=.37, P=.0001). Those who scored the highest on the knowledge instrument were also more likely to score the highest on the attitudinal inventory. When testing all independent variables, the full model was not significant [F (8,103)=1.99, P=.055).

DISCUSSION

The following conclusions may be drawn from the results of this study:

* The caregivers in this study reflect the status of child-care programs in the United States. Because of low wages and high turnover rates within the industry, caregivers are often young and inexperienced (22). In this study, caregivers had a median age of 22 years and a mean of 5 years of experience.

* Caregivers come to the classroom with varying degrees of educational preparation. The requirement and availability of nutrition training within these preparatory programs do not appear to be consistent and may not include the information that these caregivers need to successfully manage the nutrition practices for young children. General nutrition courses tend to focus on content, not on the process of incorporating nutrition into teaching practice. Little time, if any, is spent on the development of eating behaviors and nutrition practices in young children. Moreover, early childhood caregivers tend to be younger and thus have had fewer opportunities for inservice training.

Caregivers believe they influence children's eating habits, but their observed behaviors do not always support this belief. If caregivers do not sit with children, consume the same foods and create a positive, pleasant environment, then there may not be positive role modeling. The Figure provides an outline of the optimal caregiver behaviors that can be used to further clarify the role of the caregiver in nutrition practices.

* Overall, caregivers believe nutrition is important and that a child's nutritional intake will influence the long-term health of that child.

* The caregivers in this study scored low on nutrition knowledge, and the results of their performance indicate several misconceptions regarding the science and application of nutrition principles.

It is a concern that only 2 of the 24 child-care programs that were observed used a family-style meal service. Branen and Fletcher (23) found that children will consume more when allowed to self-select their intake without increasing the amount of waste.

* Mealtime was not used constructively in most of the childcare programs that were observed. A great deal of "rulesetting" took up most of the interaction between children and caregivers.

* Very little nutrition education took place during meals. When it did occur, it was usually about identification of foods and food origins.

* Nutrition knowledge and attitudes toward nutrition appear to be correlated. This will be important in helping develop training programs for early education teachers.

The results of this study begin to more clearly define the role of the caregiver in early child-care programs regarding nutrition practices. The results demonstrate that nutrition knowledge and attitudes have a relationship to how caregivers behave at mealtime; therefore, these instruments can be used to assess and teach nutrition practices through selfassessment, inservice training, and formal coursework. Through these interventions, we can influence caregiver behavior and positively affect children's eating habits. As dietetics professionals, we can use these results to be proactive in changing how nutrition is practiced in child-care programs and ultimately, ensure the sound nutritional health of our children.

This work was sponsored in part by a grant from The Center for Rural Health, Economic, and Social Development, Southern Illinois University, Carbondale.

Portions of this study were presented at The American Dietetic Association Annual Meeting and Exhibition, October 1995, Chicago, Ill.

Data are taken from a dissertation submitted by Marcia Nahikian-Nelms to Southern Illinois University in partial fulfillment of the requirements for the doctor of philosophy degree.

The author would like to thank Paul Sarvela, PhD, Sara Long Anderson, PhD, RD, Susan Pearlman, PhD, Roberta Ogletree, PhD, Kathleen Welsheimer, PhD, and Catherine Mogharreban, PhD, for their assistance with the development of this study.

Footnote:

1Requests for instruments may be made directly to the author co Department of Human Environmental Studies, Southeast Missouri State University, Cape Girardeau, MO 63701.

Reference:

References

Reference:

1. Birch LL. Experimental determinants of children's food preferences. In: Katz L, ed. Curr Top Child Ed. Norwood, NJ: Ablex; 1980.

2. Birch LL. Effects of peer models' food choices and eating behaviors on preschoolers' food preference. Child Dev.1980;51:489496.

3. Birch LL, Zimmerman S, Hind H. The influence of social-effective context on the formation of children's food preferences. Child Dev. 1980;51:856-861.

4. Birch LL, Marlin DE. I don't like it; I never tried it: effects of exposure on two-year-old children's food preferences. Appetite. 1982;3:353-360.

5. Birch LL, Marlin DW, Rotter J. Eating as the "means" activity in a contingency: effects on young children's food preference. Child Dev. 1984;55:431-439.

6. Birch LL, McPhee L, Shoba BC, Pirok E, Steinberg L. What kind of exposure reduces children's food neophobia? Appetite. 1987;9:171-178.

7. Birch LL, McPhee L, Sullivan S. Conditioned flavor preferences in young children. Physiol Behav. 1990;47:501-505. 8. Birch LL, McPhee L, Sullivan S. Conditioned preferences: young children prefer flavors associated with high dietary fat. Physiol Behav. 1990;50:1245-1251.

9. Birch LL, Johnson S, Andresen G, Peters JC, Schulte M. The variability of young children's energy intake. N Engl J Med. 1991;324:232-237.

10. Pelchat ML, Pliner P. "Try it. You'll like it." Effects of information on willingness to try novel foods. Appetite. 1995;24:253-265. 11. Drake M. Menu evaluation, nutrient intake of young children, and nutrition knowledge of menu planners in childcare centers in Missouri. J Nutr Educ. 1992;24:145-148. 12. Briley ME, Roberts-Gray C, Simpson D. Identification of factors that influence the menu at child care centers: a grounded theory approach. JAm Diet Assoc. 1994;94:276-281. 13. Food and Nutrition Board. Recommended Dietary Allowances. lOth ed. Washington, DC: National Academy Press; 1989. 14. The Food Guide Pyramid: A Guide to Daily Food Choices. Washington, DC: US Department of Agriculture, Human Nutrition Services US Government Printing Office; 1992. Home and Garden Bulletin No. 252.

15. US Department of Agriculture. Child and Adult Care Food Program Regulations. Washington, DC: US Government Printing Office; 1992.

16. Satter E. How to Get Your Child to Eat.. But Not Too Much. Palo Alto, Calif: Bull Publishing; 1987. 17. Position of The American Dietetic Association: nutrition standards for child care programs. JAm Diet Assoc. 1994;94:323-328. 18. Head Start Bureau. Head Start Program Performance Standa.rds. Washington, DC: US Dept of Health and Human Services; 1984.

19. Bredekamp S, ed. Accreditation Criteria and Procedures of National Academy of Early Childhood Programs. Washington, DC: National Association for Education of Young Children; 1992. 20. Position paper on nutrition in child care settings. J Nutr Educ. 1991;23:49-50.

21. Stanek K, Abbott D, Cramer S. Diet quality and the eating environment of preschool children. JAm Diet Assoc. 1990;90:15821584.

22. Children's Defense Fund Budget. Washington, DC:Children's Defense Fund; 1994.

23. Branen L, Fletcher J. Effects of restrictive and self-selected feeding on preschool children's food intake. J Nutr Educ. 1994;26:273-277.

Author Affiliation:

M. Nahikian-Nelms is the director of the Didactic Program in Dietetics and an assistant professor in the Department of Human Environmental Studies at Southeast Missouri State University, Cape Girardeau, MO 63701.