3 summary articles in 4 hrs EMERGENCY

Food, Culture, and Family: Exploring the Coordinated Management of Meaning

Regarding Childhood Obesity

Mozhdeh B. Bruss, Joseph R. Morris, Linda L. Dannison, and Mark P. Orbe

Department of Family and Consumer Sciences Western Michigan University

Jackie A. Quitugua Public School System

Commonwealth of the Northern Mariana Islands

Rosa T. Palacios Division of Public Health

Commonwealth of the Northern Mariana Islands

Increased rates of childhood obesity combined with more accessible information about the relationship between diet, physical activity and inactivity, and chronic dis- eases suggest the need for analyzing the complex process of receiving and transmit- ting messages related to child feeding practices. This study examined the perceptions of childhood obesity within 1 multiethnic community, the Commonwealth of the Northern Mariana Islands. In particular, through the use of focus groups, individuals indicated that sociocultural, familial, and official nutritional messages were most in- fluential to their health care behaviors. The coordinated management of meaning (CMM) theory was used to gain insight into how individuals negotiate competing messages occurring at different levels of meaning. Given its focus on cultural influ- ences (parallel to the concepts of archetypes), CMM proved especially relevant for understanding child feeding beliefs, values, attitudes, and practices in diverse ethnic populations. Implications for future health communication research that might draw from a CMM approach were identified, as well as pragmatic endeavors that focus on

HEALTH COMMUNICATION, 18(2), 155–175 Copyright © 2005, Lawrence Erlbaum Associates, Inc.

Correspondence should be addressed to Mozhdeh B. Bruss, Department of Family and Consumer Sciences, Western Michigan University, 1011 Trimpe Building, Kalamazoo, MI 49008. E-mail: [email protected]

the development, implementation, and evaluation of culturally appropriate interven- tions in the prevention of childhood obesity.

Historically, childhood obesity has been studied through the perspective of genet- ics, diet, and environmental factors. Some studies have also looked at sociocultural factors (Frideres & Goldenberg, 1982; Laroche, Kim, & Tomiuk, 1999; Powder- maker, 1997) including ethnic identity, ethnic dietary habits, consumption of eth- nic and nonethnic foods, and the cross-cultural significance of food. Recent stud- ies have explored the influence of interpersonal networks on health care decision making, including decisions related to diet and nutrition (Tardy & Hale, 1998a) and the impact that family communication has on healthy behaviors (Baranowski, Nader, Dunn, & Vanderpool, 1982; Rimal & Flora, 1998). Applying communica- tion theory, further work is needed to explore how communication within the fam- ily context is influenced by culture and how this communication may impact health-related behaviors and attitudes.

The increasing prevalence of obesity has become a social and public health con- cern with economic implications (Wolf & Colditz, 1998). Studies have shown in- creasing rates of childhood obesity among U.S. ethnic groups (Dwyer et al., 2000; Goran, 2001) with sociocultural factors playing a significant role in its develop- ment (Bruss, Morris, & Dannison, 2003; Powdermaker, 1997). These factors may be influenced by cultural perceptions regarding dietary practices (Meigs, 1997; Powdermaker, 1997). In particular, ethnic differences have been observed in di- etary intake of children (Brady, Lindquist, & Herd, 2000). There is, however, less information regarding the cultural influences in certain U.S. minority groups, par- ticularly Asian and Pacific Islanders. Food and food-related activities are a focal element in traditional Pacific Island cultures (Meigs, 1997; Powdermaker, 1997).

To study the cultural influences on dietary practices, how communication within the family context is influenced by culture, and how this communication impacts health-related behaviors and attitudes, we needed a stable multicultural community with relatively distinct cultural subgroups and high rates of obesity within these subgroups. Ideally this would include multiple ethnic groups in the United States. The Commonwealth of the Northern Mariana Islands (CNMI), United States, with a population of 62,392 (U.S. Department of Commerce, 2000) is a pluralistic society with distinct ethnic populations of Pacific Islanders and Asians and is ideal for this research. Chamorros and Carolinians are indigenous to the CNMI, whereas other Micronesians (Palauans and Chuukese), Filipinos, Kore- ans, Japanese, and Chinese, are recent immigrants. Obesity is highly prevalent in the CNMI with childhood obesity a growing public health concern (CNMI Food and Nutrition Advisory Council, 1996).

In understanding obesity, it is important to recognize the role that culture plays in the negotiation of multiple messages. Cultures communicate information re-

156 BRUSS, MORRIS, DANNISON, ORBE

garding child-feeding practices in terms of beliefs, values, and behaviors related to different foods. In recent times nutrition education has also communicated child- feeding information to caregivers, specifically in regard to the prevention of child- hood obesity. A caregiver’s perception of appropriate child feeding practice is in- fluenced by a variety of sources, including culture, family characteristics, social experience, and environmental communicative messages. The relative impact of information from these sources—and the weight that is given to each—is depend- ent on family needs, attitudes, values, and goals (Bubolz & Sontag, 1993). In studying the role of culture on the development of dietary habits within the family context, the use of qualitative research may be the best approach (Orbe, 2000). This methodological approach will provide a better understanding of the relation- ship between child feeding practices and childhood obesity, as well as guide the design of effective interventions. This study focused on family messages related to child feeding. Specifically, the study examined how particular messages were re- flective of parents’ beliefs, values, attitudes, and practices across different cultures and how these messages were negotiated among those that were communicated through sociocultural and more formal channels. We used an interdisciplinary team of scholars and practitioners from family and consumer sciences, counselor education/counseling psychology, communication, health promotion, and educa- tion to study how family members communicate dietary habits.

CONCEPTUAL AND THEORETICAL FOUNDATIONS

The coordinated management of meaning (CMM) theory is a body of work that provides a framework to structure our themes, albeit in a post-design consider- ation. In this regard, our exploratory qualitative study is not a test of CMM but in- stead an illustration of its utility for framing our findings. Although we draw from research across disciplines, research specifically in health communication pro- vides a foundation for our study. First, through the advancement of a health belief model together with other conceptual frameworks, research has identified basic factors that influence preventive behaviors (e.g., Chew, Palmer, & Kim, 1998; Chew, Palmer, Slonska, & Subbiah, 2002). Specifically, the health belief model fo- cuses on preventive health care behavior as a function of readiness to act and effi- cacy of the recommended response (Chew et al., 1998). Signorielli and colleagues (Signorielli & Lears, 1992; Signorielli & Staples, 1997) have demonstrated that media also play a significant role in determining children’s perceptions of what constitutes healthy eating. Second, recent scholars have called for health commu- nication that focuses on everyday interpersonal communication, something that Cline refers to as the “missing box” in health communication literature (Cline, 2003). Consequently, this research moves away from an exclusive emphasis on tra- ditional foci in health communication (i.e., doctor’s communication, health cam-

CHILDHOOD OBESITY 157

paigns) and concentrates on studying the “impact of informal, interpersonal net- works in health care decision making” (Tardy & Hale, 1998a, p. 151). It adds to other work that examined the impact of family communication on healthy behav- iors (Baranowski, Nader, Dunn, & Vanderpool, 1982; Rimal & Flora, 1998), as well as studies focused on informal support groups of nonrelated individuals (Albrecht & Goldsmith, 2003; Bauer & Orbe, 2001; Tardy & Hale, 1998b). Based on these studies, health communication researchers have gained valuable insight into differences in how individuals and families frame “noncompliance” and “healthy” behaviors (Babrow & Mattson, 2003; Mishler, 1984). Given the influ- ence of culture on behavior (Coltrane, 1998; Cronk, 1999), understanding how families communicate within a cultural context is crucial to ultimately facilitating effective health practices. Consequently, this study is grounded in a sociocultural tradition of health communication research (Craig, 1999).

Sociocultural, genetic, and familial factors interact to influence human behav- ior generally and human communication specifically. As such, recent works have focused on how family dynamics, situated within particular cultural contexts, im- pact communication (Socha & Diggs, 1999). Because families exist within a larger social context, they are shaped by a range of variables that contribute to their diver- sity. Ethnicity, historical influences, geography, and socioeconomic status are ex- amples of factors that influence families (McGoldrick, Giordano, & Pearce, 1996). By interacting with external environments, families adapt and transmit beliefs, at- titudes, and values, which shape the behavior of individual family members, a pro- cess which is intergenerational by nature (Bowen, 1978; Cox & Paley, 1997; Goldenberg & Goldenberg, 1999; Papero, 1990).

Originally introduced in the mid-1970s (Pearce, 1976), the CMM is grounded in a basic premise: Communication is the process by which people “co-create, maintain, and alter social order, personal relationships, and individual identities” (Cronen, Pearce, & Harris, 1982, p. 64). The CMM theory contends that people co- ordinate their lives by managing the ways in which messages have meaning for, and through, larger patterns of meaning (Heath & Bryant, 2000). Pearce and Cronen (1980) describe communication as “a process in which each person inter- prets and responds to the acts of another, monitors the sequence, and compares it to his or her desires and expectations” (p. 68). Consequently, CMM depends on par- ticular interaction rules, the content of messages, and the ways in which various kinds of interaction are structured (Heath & Bryant, 2000). The CMM theory sheds insight into the process of meaning construction and, in particular, the ways that individuals create meaning among multiple interpretive levels (coherence) to- ward some specific outcome (coordination).

Families receive multiple, sometimes competing, messages regarding nutrition and dietary habits. CMM was used to understand how people use complex, multileveled systems of rules to derive meaning and guide behaviors (Pearce &

158 BRUSS, MORRIS, DANNISON, ORBE

Cronen, 1980). People interpret messages and know what actions constitute appro- priate responses because they can follow rules that guide what they do and say (regulative rules) and how they interpret what transpires (constitutive rules). Rules help individuals know what behavior is appropriate and productive (regulative), and how to interpret other’s behavior (constitutive; Cronen et al., 1982).

According to CMM theory, people learn behaviors that are appropriate for spe- cific contexts (Pearce, 1976). In particular, they rely on an interpretative process that includes six levels of understanding (Pearce & Cronen, 1980; Philipsen, 1995). These are: (a) content (the words used to communicate), (b) speech acts (how we perform the content), (c) contracts (a system of formal and/or informal rules that guide two or more individuals’ communication), (d) episodes (communi- cation routines that consist of a describable sequence of speech acts), (e) life scripts (individual’s self-perception that shapes, and is shaped by, communica- tion), and (f) archetypes (understandings of speech acts, contracts, episodes, and life scripts that are shared by a particular social group). Consistent with the per- spective taken within this study, CMM situates archetypes, described by commu- nication scholars as cultural patterns (Wood, 1997), in the fundamental logic that people use to frame or define experience (Pearce & Cronen, 1980). In short, CMM provides a framework for the primary objective of our study: to explore the process through which individuals in a multiethnic community negotiate multiple mes- sages regarding childhood obesity.

METHODS

Study Design and Participants

This study was designed in consultation with two local collaborators who facili- tated a culturally appropriate protocol. The study used qualitative methods in two ways. First, one of the authors worked as a public health nutritionist in the CNMI and had familiarity with the community, its cultural groups, and its nutrition issues. In this regard, the project utilized ethnographic methods, an approach often used by both health communication researchers (e.g., Tardy & Hale, 1998) and CMM scholars (Philipsen, 1995). Second, four focus groups were conducted in February 2002, in the CNMI (total N = 32). Participants included mothers, fathers, and grandparents of children 6 to 10 years old from four public and private schools in Saipan, selected based on the predominant ethnic makeup of the student body (Chamorro, Carolinian, and Filipino). Participants self-reported ethnicity and age of children in the household. The Western Michigan University Human Subject Review Board approved the research proposal and informed consent of the partici- pants was obtained.

CHILDHOOD OBESITY 159

Data Collection

Three trained moderators used a semistructured interview protocol in English (the common spoken language). Sessions were between 60 and 90 min in length and designed to gain information regarding how individuals came to define, and under- stand, issues related to nutrition, dietary habits, and childhood obesity. Participants were invited to answer open-ended questions related to their personal experience, their children’s experience, and communication that may influence child feeding and weight status among 6- to 10-year-old children.

Interviews were audiotaped and transcribed by a Saipan-based transcription agency that is familiar with accents of the different groups in the CNMI. Content of each transcript was verified by the first author for accuracy against audio tapes. Data analysis was conducted using methods previously described (Bogden & Biklen, 1998). The first three authors read the transcripts and developed a system of coding that involved developing coding categories, assigning each code a num- ber, marking each comment with appropriate number(s), and adding new codes as needed. Transcripts were coded independently and compared for interrater reli- ability. Individual coding that was not in accordance with all three researchers was discussed until a consensus was reached. Once categories were established and il- lustrated through participant comments, the fourth author reframed them using a CMM theoretical framework. Although results of this inquiry cannot be general- ized, the multiethnic composition of the focus groups strengthens potential for applicability.

RESULTS

Based on our analysis, it became abundantly clear that individuals in families ne- gotiate meanings regarding nutrition within the context of different message sources. Specifically, three major interrelated types of messages emerged from the data as most significant: (a) sociocultural messages (grounded in history, culture, and social experiences), (b) familial messages (reflective of family characteristics, needs, values, attitudes, and goals), and (c) nutritional messages (inclusive of nu- trition information from different official sources). We describe the content of these messages and discuss the meanings within CMM’s six levels of hierarchical understanding.

Sociocultural Messages

Sociocultural messages reflect historical influences and social experiences im- pacting caregivers’ perceptions of child feeding. It is within informal forms of communication that messages are sent regarding particular attitudes, beliefs, and

160 BRUSS, MORRIS, DANNISON, ORBE

values placed on certain foods and eating patterns. Attitudes toward eating habits and foods such as rice, fresh foods versus canned foods, vegetables, and foods with added sugar are communicated through social gatherings as well as interaction among caregivers and the general population. Historically, rice in the Mariana Is- lands, which dates back to 1,500 BC, has played a crucial role in dietary practices. In fact, rice occupies a preeminent position in contemporary CNMI society, as ex- pressed by one father:

My grandma, my mom, everybody tells me you’ve got to eat rice, no limit. You are also encouraged to have second portions of it. If you finished and you asked for another one, they say “go for it.”

In a similar manner, caregivers’ preference for fresh food versus canned foods is rooted in island culture, which promoted use of readily available foods from the land and sea. As one participant stated, “You can’t survive without local foods; they are the healthy food for local people. We try to avoid canned foods.” Among the Pacific Island population, thinness has a more negative connotation especially when it affects children (Bruss, Morris, & Dannison, 2003). Sociocultural attitudes related to vegetables suggest that caregivers associate eating vegetables with thin- ness and believe that excessive amounts of vegetables are not desirable. This view may be rooted in the belief that vegetable consumption is related to weight status. One participant explained that “sometimes too many greens are not too good ei- ther.” Another shared that:

My mother says my 6-year-old son is really, really thin. He only likes greens and vegetables. He does not like anything else. We try to force him to eat and it is hard.

Within this example, the hierarchical nature of coordinated meaning is apparent. The grandmother’s message is clear (content), given the greater power of the rela- tionship between mother and daughter (speech act). This short quote, and the con- text in which it was shared, also indicates that the family has struggled with negoti- ating the son’s and parents’ desires and expectations (contracts). This has been something that has continued over time (episodes), resulting in competing life scripts for the parents and the child. Most central to this section, however, is the overarching influence of cultural archetypes that define the son as thin.

A similar dynamic was seen in how parents discussed recent dietary changes, including increased access to discretionary sweets. In response to changes, care- givers have adopted strategies (contracts) to communicate acceptance of foods through practices, such as “payday treats,” which is when parents offer children food treats on paydays. However, they indicated that they do not endorse the

CHILDHOOD OBESITY 161

hoarding of sweets even on holidays such as Halloween and identified the chal- lenge with their availability in the community and at social gatherings.

When we were growing up, it’s really hard to buy candy, and that’s if you can get it.

I have to be really firm about it, but when it comes to Valentine or Hallow- een, there is candies everywhere; it is hard.

When we come to parties, my kids will stay in front of me, and ask why those other people can have so many, and I say, “yes, you can have some but not the whole can.”

Sociocultural messages within the community also contained clear guidelines (ar- chetypes) as to what were considered acceptable eating habits. Yet parents used different strategies, consistent with their own life scripts, to evaluate their chil- dren’s dietary intake. In addition to using the child’s intake as a standard for deter- mining individual eating habits, caregivers reported comparing their children’s eating habits with other children. For instance, one explained,

I think my kids eat more than they should compared to other kids, and I’m still trying to work at it, trying to limit what they eat, and if they do eat more, I’ll try and make them do more activities. They eat more than they should.

We can see that some of what parents perceived as acceptable eating habits for their children was determined by the local community, seemingly the most influ- ential of CMM’s hierarchy. However, individual families and their collective life scripts also served as an influence in creating meaning regarding eating norms.

Familial Messages

Familial messages are another form of communication that influences caregivers’ perceptions of child feeding. Often these messages were consistent with those con- tained in sociocultural messages. Especially interesting in our findings was the way some familial messages—reflective of a set of collective contracts, episodes, and life scripts—were negotiated against larger archetypes regarding childhood eating. Through data collection, we learned different ways families communicate (implicitly and explicitly) value statements regarding food consumption. As fami- lies struggle to define needs and values related to child feeding, they seem to have specific views about foods that are acceptable and those that are to be avoided. As mentioned earlier, some of these were rooted in a larger cultural context. However, others seem specific to individual family life scripts.

162 BRUSS, MORRIS, DANNISON, ORBE

For instance, caregivers’ definition and attitudes (individual life scripts) related to snack foods varied. These included avoidance of snack products; occasional of- fering of foods such as cereal, so long as it did not interfere with meals, especially rice; purchasing candy and chips to please the children; and frequent offering of fruits. The following focus group excerpts illustrate how individual life scripts can exist within one community.

If you avoid snacks, sweets, and fatty foods then that’s good; you can eat anything as long as you avoid all those things.

My kids, they must eat chocolate. If my husband goes to the store, he buys candy for my kids. They like candy and chips.

In general, fruits were more acceptable than vegetables, especially as snack foods. Caregivers identified fruits as snacks, showed preference for fruits over chips, and considered fruits to be part of a healthy eating pattern. Overall, a positive attitude was observed in regard to fruits: “They can eat it [fruit] in every meal and use it as snack; it’s how much they can eat.”

In regard to added sugar, we found that it is common practice to add extra sugar to foods that are already sweetened, such as sweet powdered drink mixes and cara- mel apples. Throughout focus group discussions, this episodic negotiation of a larger archetype was described by many as natural (i.e., “We go ahead and put sugar again.”). However, fathers pointed to particular contracts that they had adopted that limited intake of such foods to one candy bar, soft drink, or a similar item per day. They agreed that children should be limited to one or two of these items per day, recognizing that if the parent offered one of the items, the child would obtain a second one from another source.

In regard to beverage consumption, caregivers reported to read food labels and purchase 100% fruit juice despite higher cost. However, caregivers indicated that in reality a certain brand of sweetened orange drink is more commonly offered al- though parents recognize beverages need to be 100% fruit juice. Other parents in- dicated that they preferred not to purchase any sweetened beverages to avoid con- flict with children: “No sodas in the house because every time you buy them, they even want it in the morning; they want to open and drink it.”

With increased prevalence of chronic diseases, the content of health care mes- sages in the CNMI has focused on role of fat in the diet. Caregivers did not express a concern about eating foods with added fat, especially when children are per- ceived as normal weight: “My kids are not fat, they are not very thin, they are just normal, and so I fry chicken and eggs. I fry most of the time, but they seem to be okay.” This quote illustrates ways in which content of health care messages was ne- gotiated against larger conceptions of meaning regarding weight and eating habits. This type of scenario, in which meaning was coordinated among divergent mes-

CHILDHOOD OBESITY 163

sages, was varied for different groups. Among fathers, there was an information gap related to identifying strategies that limit such foods. Filipino participants identified canned meats as foods preferred by children and a source of fat in the diet (episodes). However, differing views in regards to which canned foods were higher in fat were observed (life scripts).

Among participants, family life scripts regarding caregiving and food were clearly observed. Study participants indicated that in Micronesia parents demon- strate love for a child by “holding the child’s feelings,” which in part means giving the children what they want. Parents struggle with trying to strictly maintain a healthy diet and accommodating their children’s preferences. Caregivers reported that eating episodes were easier when they prepared foods that they know their children will eat. They also find it difficult when their children do not eat their food. In fact, children’s overall consumption seemed to be of concern to focus group participants. Caregivers identified their children as picky and difficult to please, and reported having to prepare foods in a manner that would increase their consumption. In general, caregivers believed that children control their own eat- ing, despite parental concern with overall food consumption and potential weight loss. Challenged by this issue, caregivers find themselves at a loss for an appropri- ate strategy to deal with their children’s feeding concerns.

When you cook soup, they comment, but when it is lightly fried they tend to eat. If it is soup, you can see their face, they are so picky, the kids now days are more picky.

Intergenerational family influences were also noted, with mothers reporting the influence of their mothers on child feeding practices. This phenomenon helps to illustrate the inextricable relationship between cultural archetypes and life scripts and the ways that lower levels of meaning are cocreated within a larger context. Mothers indicated personal experience with food and individual food choices as factors influencing their child feeding beliefs and practices. This involved both restrictive (“For me no, and for my child, no, I don’t buy. They don’t see me eating candies or drinking soda. No, I don’t eat and I don’t buy candy or chocolate”) and nonrestrictive practices (“I do keep sweets in the house because I love chocolates”).

Within each caregiver’s description of their families’ life script, they identified children’s preferences for foods as a major factor in determining their diet and how speech acts regarding eating were performed. Regarding this issue, parents also expressed the belief that individual children in general may have different prefer- ences for their weight status and food intake, resulting in the need for particular contracts with each child.

164 BRUSS, MORRIS, DANNISON, ORBE

Mine is different, because one is into sweets and the other one is not really into sweets.

It all depends on the child. One child hates to be fat, the other one likes being fat. It all depends on the child. “I don’t want to be fat like him.” “I don’t want to eat that.“

Caregivers also reflected on children’s individual preferences for certain foods, such as meats over vegetables and chicken skin or a general dislike for lard, may- onnaise, and chicken. Most mothers identified canned meats as a favorite food with a general dislike for vegetables.

The kids like fried canned meat. If you cook vegetables, they don’t like it. They would rather eat cereal than vegetables. Or open the can of meat and that’s the only time they can eat good.

As described, many familial messages were inextricably linked to those operat- ing within a larger sociocultural context. It was most evident when focus group participants discussed traditional food preparation methods and the cultural values with which they were associated. Undeniable acceptance of traditional food prac- tices, such as frying, is passed down through generations, as reflected in this fa- ther’s statement, “My grandma ate fried fish, then my mom, and now me. I think there is nothing wrong about that.” However, some meanings grounded in arche- types are challenged by individual and family life scripts. This is important to rec- ognize when constructing health care messages whose meanings appear to contra- dict cultural norms. The effectiveness of nutrition messages that do not consider the process of meaning negotiation, as seen in the following section, is signifi- cantly reduced.

Nutritional messages

Throughout focus group sessions, participants also discussed nutritional messages that they received from formal sources. These messages were received and inter- preted alongside those that were gained more informally (e.g., sociocultural and familial messages). For example, in addition to transmission of sociocultural and familial messages about rice, issues were raised by participants that indicated im- pact of nutritional messages on caregivers’ understanding of rice in child feeding situations. Recognizing the value of rice in the diet, nutritional messages in the CNMI have focused on type, frequency, and amount of rice intake, along with sub- stitution with roots and tubers. Caregivers had varying views on the appropriate amount of rice intake: one scoop at every meal, one scoop per day, one cup per day,

CHILDHOOD OBESITY 165

one to two cups at each meal, or three cups per day. In one focus group discussion, mothers discussed serving utensil and portion size before and after cooking, and they tried to negotiate an appropriate amount given how meaning was negotiated within episodes, life scripts, and archetypes. The group consensus was one-half to one cup of rice twice daily.

In an attempt to assess caregivers’ perceptions of foods with added sugar, par- ticipants were asked to estimate the appropriate intake for 6- to 10-year-old chil- dren. Parents reported various forms of contract that they implemented, includ- ing avoiding purchasing sweets, candies or similar foods as a measure of preventive dentistry. Parents who perceived their children as normal weight sta- tus, according to life scripts and cultural archetypes, reported that they did not see a relationship between sweets and weight status. However, some indicated an association between hyperactivity and poor listening with intake of sweets, saying that “We stopped buying sweets” or “I don’t give sweets to them.” These strategies appeared to produce particular contracts in light of ongoing episodes between parent and child.

In general, caregivers identified schools and medical professionals as sources of information for appropriate child feeding practices. However, caregivers’ under- standings and interpretations of nutritional messages were varied, contingent on earlier episodes with these individuals. The belief that the schools know what is nutritious and offer daily fruit juice to the children is an indication to parents that this was an appropriate practice. Apple and orange juices were considered to be fruit juices and therefore appropriate for children, but soft drinks were not consid- ered appropriate. The communication/interpretation of dietary recommendations in regard to regular soft drinks is reflected in this mother’s statement, one that illus- trates a negotiated contract:

I rather he drinks one diet soda a week than one regular soda. If he has a treat, it is usually diet soda. I could let him have a candy bar, but he has a diet soda. And, when sometimes he does not eat his food, he knows the ground rule; he will be out of luck with the soda pop.

Across the focus groups it became readily apparent that participants negotiated the meaning of nutrition messages within the context of cultural archetypes and in- dividual life scripts. For example, caregivers’ perceptions of healthy child feeding habits varied considerably. Eating healthy meant different things, including eating three meals a day; having fruits and vegetables as a snack instead of chips; having a chamorro merienda (afternoon snack); consuming the right amount of food; a diet that includes the basic food groups and has variety; eating locally grown foods and vegetables, following the food pyramid; and avoiding snacks, sweets, and fatty foods. Caregivers, in general, agreed that limiting sweets is part of developing healthy eating habits, which needs to begin at an early age. One described how

166 BRUSS, MORRIS, DANNISON, ORBE

consistency in establishing clear episodes was important, and said, “I think it is good to start it from when they are young … start at 18 months, not when they are 18 years old; that’s too late already.” Most caregivers reported how nutritional messages prompted a conscious restriction of these foods, identifying fruits and vegetables as appropriate substitutions. Among fathers, access to such foods in the house was seen as tempting, hence the preventive measure to withhold their entry into the home. Yet implementing this increased awareness appeared difficult given that life scripts, established over years of episodes, were incongruent with sug- gested restrictions.

One example of this difficulty involved selecting foods lower in fat. Based on formal nutritional messages, caregivers also identified reading labels as a strategy to distinguish foods that are lower in fat. However, they indicated that this was not commonly used by shoppers, who look for lower cost items, a salient aspect of many participants’ life scripts. “Normally, that’s what we do, look for the cheapest; we don’t think about how many grams of fats it has, or is it really a nutritious food or not,” said one participant. Other dietary practices reported by caregivers related to reducing fat content of foods. Specifically, these included removing fat before cooking, boiling food and then removing the fat, and steaming instead of frying. In addition, other strategies were reported by focus group participants concerning canned meats. These included limiting purchase of the meats, offering them weekly instead of daily, and serving them occasionally in the mornings when care- givers don’t have time to cook breakfast. Parents also identified lower fat options such as fish for this purpose. Related to this, most fathers recognized the need for dietary modification, “We just have to change the way we cook our food.” How- ever, this was different given the value put on traditional cooking practices re- flected in larger cultural archetypes.

DISCUSSION

In this article we reported three interrelated themes that do not exist in isolation; rather, they exist in a complex web of interwoven factors that influence caregivers’ perceptions of child feeding. Through a CMM theoretical framework we can un- derstand how the meaning of what constitutes acceptable eating habits, in terms of particular constitutive and regulative rules, is cocreated along six levels (content, speech acts, contracts, episodes, life scripts, and archetypes). Through the discus- sion of three different types of messages, we have provided examples that reflect each of these levels. However, given the hierarchical nature of CMM theory we turn to the overarching influence that archetypes (or cultural values) play in terms of negotiating multiple messages regarding dietary habits. For our participants, culture serves as a “logical force,” “a deontic logic implicating what they should, must, may, or cannot do in specific situations” (Pearce & Pearce, 2000, p. 416), in

CHILDHOOD OBESITY 167

decision making regarding food consumption. As such, it warrants greater explo- ration for both health care practitioners and researchers.

Implications for Praxis

The influence of sociocultural cues on health communication practices must be recognized for practitioners to effectively advocate for change (Airhihenbuwa, 1995; Ford & Yep, 2003; Geist, 1994). These findings suggest that sociocultural messages, those communicated within local communities and reflected in histori- cal influences and social experiences, appeared to significantly impact caregivers’ perceptions of child feeding. Our analysis indicated that participants’ communica- tion regarding food contained clear connotation regarding particular values placed on certain foods and eating patterns. For example, Pacific Island participants seemed to value rice, local foods, fresh food, and traditional meal patterns that in- fluenced their child feeding practices. Previous studies support the notion that his- torical and sociocultural factors related to food may be influencing the indigenous cultures’ responses to current dietary practices (Meigs, 1997; Powdermaker, 1997). Food, for example, plays a major role in family gatherings and socialization (Reidy, Weinstein, Milgrom, & Bruss, 2001), something that was noted in CNMI by investigators. Implicit in these events are sociocultural messages to caregivers about the importance of generous food offerings and abundant hospitality in the culture. These settings may pose a challenge for the caregiver, especially given the ubiquitous nature of sociocultural values in everyday life.

According to our findings and consistent with CMM, nutritional messages need to take into consideration the sociocultural attitudes related to several key values, including those related to “weight normalcy,” defined as acceptable body image, and traditional cooking practices. Caregivers’ attitudes toward discretionary sweets were less varied and suggested active parental transmission of messages related to such foods. However, an ongoing concern is the ability of caregivers to control children’s intake of discretionary sweets (via contracts) when away from the home environment and in other social settings.

Caregivers used different levels of understanding to cocreate meaning regard- ing their children’s eating habits. In some cases caregivers used the child’s intake as a standard for determining individual eating habits (episodes), and in other cases they were inclined to compare their children’s eating habits with other children (re- flected in larger archetypes). With an increasing concern for childhood obesity (CNMI Food and Nutrition Council, 1996), community-based parent education programs might be more successful if the larger cultural values related to child feeding practices are considered.

Educators must recognize that larger sociocultural influences (archetypes) help to cocreate meaning alongside family messages (life scripts). For instance, while caregivers’attitudes toward vegetables were influenced by archetypes, they reported

168 BRUSS, MORRIS, DANNISON, ORBE

buying vegetables along with fruits as a strategy to promote healthy eating for their children (family-based scripts). Although caregivers may experience conflict over divergent messages, in some cases they seem to make independent decisions that re- flect their circumstances and individual life scripts. For instance, caregivers’ atti- tudes related to snack foods were varied. A number of participants communicated their concern with the interference of snack foods in children’s meal patterns. Others identified the reinforcement of children’s preferences for such foods by family members as the most salient level for meaning creation. In general, caregivers of- fered different definitions for sweets and sweet foods, which could hinder accep- tance of nutrition messages. Clarity of definitions and use of culturally sensitive lan- guage in nutrition education messages related to child feeding may require qualitative studies across demographic groups based on age, gender, and/or culture.

Nutritional messages specifically related to rice, discretionary sweets, added fats, and healthy child feeding habits were found to influence caregivers’ percep- tions of child feeding beliefs, values, and practices. Still, caregivers reported their current use of discretionary sweets, which included the common practice of adding extra sugar to foods that are already sweetened. In tropical climates, bev- erage consumption is an important issue, especially for parents interested in of- fering suitable options for their family. Results of a study by Brady, Lindquist and Herd (2000) suggest that 7- to 14-year-old children’s dietary intake had ex- cessive amounts of energy from foods with added sugar and discretionary fats. Although caregivers recognize the value of limiting such foods, our findings in- dicate that they may find implementation challenging.

Caregivers did not see a concern for fats and fatty foods, especially when chil- dren are perceived as normal weight. Cardiovascular risk factors associated with excessive intake of such foods (Klesges, Eck, & Shelton, 1995) suggest the need for culturally relevant messages that consider caregivers’ ideas about weight nor- malcy and fat consumption. Parents communicated desire to please their children through food. Although they are aware that frying increases dietary fat, they choose this option to enhance children’s food consumption. For example, parents’ decisions to fry food to increase food acceptance suggests the important role food plays in caregiving practices. Caregivers reported that they usually prepare foods that they know their children will eat because they find it difficult when their chil- dren do not eat their food. Among the Pacific Islanders, caregivers believe one’s diet is a source of strength and illness prevention. For this reason, they feed their children to ensure weight gain, to remain healthy and to have strength. They ac- knowledged that their children are picky and difficult to please, and reported hav- ing to prepare foods in a manner that would accommodate their preferences. Care- givers believed that children control their own eating despite parental concerns with overall food consumption and potential weight loss. Challenged by this issue, caregivers find themselves at a loss for appropriate strategies to deal with their children’s feeding concerns.

CHILDHOOD OBESITY 169

Participants identified children’s food preferences, parents’ personal experi- ence with specific foods, and sociocultural messages as factors that influence children’s dietary habits. Rice consumption provides a specific example for this point of analysis. Although the widely distributed CNMI Food Guide Pyramid, which is based on the previous USDA Food Guide Pyramid, recommends 6 to 11 servings of a variety of grains, grain products, roots, and tubers, caregivers reported varying views and practices with regard to rice intake. The cultural value of rice as an integral part of the diet seems to be currently challenged by the general population-based nutrition education messages, which advocate for variety and portion control. Caregivers sought dietary guidance on rice, perhaps attempting to align cultural views with recommended nutrition information.

Dietary recommendations and advice for the general public in both the United States and the CNMI have focused on limiting discretionary sweets, which in- cludes sweets, beverages, and foods with added sugar. In regard to these foods, participants focused their responses on amount of sweets and frequency of intake. In some cases they suggested offering children substitutes such as water instead of soft drinks and sweetened drinks. Caregivers also identified the importance of lim- iting such foods as a measure of preventive dentistry. This finding can be important in the design of nutrition education interventions aimed at the prevention of child- hood obesity, specifically in regard to the use of the Food Guide Pyramid and limit- ing discretionary sweets.

Caregivers’ understandings and interpretations of nutritional messages in re- gard to these foods seemed to be varied. Caregivers identified the school as a credi- ble source of nutrition information in regard to offering juices as part of the school meal service. Caregivers in general agreed that limiting foods with added sugar is important to developing healthy eating habits, which needs to begin at an early age. Most reported a conscious restriction of these foods and the use of fruits as substi- tutes. In a study by Gabel and Lutz (2000), authoritarian-style parenting was posi- tively associated with more frequent availability of sweets in the home. Although we did not study parenting style, we found that caregivers saw an active role for parents in limiting children’s access to such foods.

Reading labels was identified by caregivers as a strategy to select foods that are lower in fat but is not commonly used by shoppers who tend to look for lower cost items. Caregivers shared a number of possible strategies to help reduce fat content of the diet. However, issues inherent to individual life scripts, such as economics, time, and insufficient information, were observed to negatively impact practice in regard to food selection and preparation. These findings support the need for par- ent education programs in the prevention of childhood obesity and reducing risk factors for coronary heart disease. Other studies have found that parent education was an effective strategy in an intervention study that aimed at reducing children’s exposure to known nutrition risk factors for coronary heart diseases (Rasanen et al., 2002).

170 BRUSS, MORRIS, DANNISON, ORBE

Findings related to participants’ interpretations of messages regarding foods and dietary practices suggest varied interpretations and levels of adherence to di- etary recommendations. Focus group sessions offered an effective forum for dis- cussions related to child feeding, especially when participants voluntarily engaged in negotiating appropriate dietary substitutes and amounts of food. The process of communicating with peer group members, facilitated by culturally competent nu- trition educators, appeared productive in guiding social/cultural constructions of acceptable dietary habits.

Implications for Future Study

This study examined the central role of sociocultural and familial messages on eat- ing, which has many implications for health care practitioners whose work is to guide healthy dietary habits. This area also provides a rich point of analysis for scholars interested in exploring issues central to health, family, and cultural com- munication. Through this study we found a number of issues, identified in the pre- vious section, that should be studied further. To avoid unnecessary redundancy, we will not repeat those here. We will therefore highlight specific implications for those researchers who may have an interest in using CMM in their research.

CMM, and its attempt to create a theoretical framework to understand how in- terpersonal communication is used to cocreate meaning at multiple levels, is not without its critics. For example, Brenders (1987) questions the role that CMM has had in advancing a “philosophically sound concept of meaning” (p. 329). Despite the existence of such criticisms, CMM emerged within our study as a valuable source for concepts that practitioners could understand and use to increase effec- tiveness of nutrition messages. This move from the theoretical to the practical is consistent with the recent work of CMM scholars, who have extended its interpre- tative beginnings to include greater practical applications (Pearce & Pearce, 2000, 2001). Specifically, these researchers have used a CMM theoretical framework to facilitate dialogue regarding social issues (i.e., multiculturalism) within various community contexts. It appears that existing structures for such deliberations al- ready exist. (See the SHEDD model as described by Pearce & Pearce, 2001. SHEDD is an acronyn for the five phases of a public dialogue process model: [a] starting, [b] hearing all the voices, [c] enriching the conversation, [d] deliberating the options, and [e] deciding how to move forward together.) Replicating this work within an area of health communication would generate valuable insight as to its applicability across contexts.

These findings will be useful for those parties responsible for the creation, im- plementation, and evaluation of nutrition messages aimed at culturally diverse populations such as those that reside in CNMI. Historically, a significant amount of research has been completed that examines the effects of various media forms on nutrition awareness and eating habits (Cornell, Goldberg, & Folta, 2001; Kava,

CHILDHOOD OBESITY 171

Meister, Whelan, Lukachko, & Mirabile, 2002; Larson, 1991). More scholars (e.g., Tardy & Hale, 1998a, 1998b) have recently begun to explore how interper- sonal communication messages within the context of family and friends affect the cocreation of meaning regarding health issues. However, no research that looks at how media influences are informed by sociocultural relationships were found. Current scholarship would benefit greatly from such research questions (e.g., How do individuals within families cocreate individual and collective meaning from particular nutritional messages communicated through various media forms?). Al- though providing a more holistic approach to studying multidimensional ways in which health messages are negotiated, this line of research can also be productive in extending current CMM research to include the influence of media in the cocreation of meaning.

In summary, many factors influence caregivers’ perceptions of child feeding practices. Sociocultural, familial, and nutritional messages contain metamessages regarding specific beliefs, attitudes, values, and practices related to child feeding, some of which are incongruent. Cross-cultural, familial, and individual differences suggest that nutrition intervention messages that are designed for the prevention of childhood obesity should consider the various levels through which meaning is cocreated. The prevention of childhood obesity, with special attention to socio- cultural and familial messages, is especially important because recent studies have found that childhood obesity is particularly intransigent in U.S. ethnic minority populations (Dwyer et al., 2000; Goran, 2001). With the cost of obesity in the United States estimated at $99.2 billion annually, the importance of preventing childhood obesity as a benefit to individuals, families, and society cannot be over- emphasized (Wolf & Colditz, 1998).

ACKNOWLEDGMENTS

The authors wish to acknowledge the support of the Diether H. Haenicke Institute for International and Area Studies, Ms. Frances T. Demapan, the CNMI Public School System, and the CNMI Department of Public Health.

REFERENCES

Airhihenbuwa, C. O. (1995). Health and culture. Beyond the Western paradigm. Thousand Oaks, CA: Sage.

Albrecht, T. L., & Goldsmith, D. J. (2003). Social support, social networks, and health. In. T. Thomp- son, A. Dorsey, K. Miller, & R. Parrott (Eds.), Handbook of health communication (pp. 263–284). Mahwah, NJ: Lawrence Erlbaum Associates, Inc.

172 BRUSS, MORRIS, DANNISON, ORBE

Babrow, A. S., & Mattson, M. (2003). Theorizing about health communication. In. T. Thompson, A. Dorsey, K. Miller, & R. Parrott (Eds.), Handbook of health communication (pp. 35–62). Mahwah, NJ: Lawrence Erlbaum Associates, Inc.

Baranowski, T., Nader, P. R., Dunn, K., & Vanderpool, N. A. (1982). Family self-help: Promoting changes in health behavior. Journal of Communication, 32, 161–172.

Bauer, K., & Orbe, M. (2001). Networking, coping, and communication about a medical crisis: A phenomenological inquiry of transplant recipient communication. Health Communication, 13, 141–161.

Bogden, R. C., & Biklen, S. K. (1998). Qualitative research for education: An introduction to theory and methods. Boston: Allyn & Bacon.

Brady, L. M., Lindquist, C. H., & Herd, S. L. (2000). Comparison of children’s dietary intake patterns with US dietary guidelines. The British Journal of Nutrition 84, 361–367.

Bowen M. (1978). Family therapy in clinical practice. New York: Aronson. Brenders, D. A. (1987). Fallacies in the coordinated management of meaning: A philosophy of lan-

guage critique of the hierarchical organization of coherent conversation and related theory. Quarterly Journal of Speech, 73, 329–348.

Bruss, M., Morris, J., & Dannison, L. (2003). Prevention of childhood obesity: Sociocultural and famil- ial factors. Journal of American Dietetic Association, 103,1042–1045.

Bubolz, M. M., & Sontag, S. M. (1993). Human ecology theory. In P. G. Boss, W. J. Doherty, R. LaRossa, W. R. Schumm, & S. K. Steinmetz (Eds.), Sourcebook of family theories and methods: A contextual approach (pp. 419–444). New York: Plenum.

Chew, F., Palmer, S., & Kim, S. (1998). Testing the influence of the health belief model and a television program on nutrition behavior. Health Communication, 10, 227–245.

Chew, F., Palmer, S., Slonska, Z., & Subbiah, K. (2002). Enhancing health knowledge, health beliefs, and health behavior in Poland through a health promoting television program series. Journal of Health Communication, 7, 179–196.

Cline, R. J. W. (2003). Everyday interpersonal communication and health. In. T. Thompson, A. Dorsey, K. Miller, & R. Parrott (Eds.), Handbook of health communication (pp. 285–318). Mahwah, NJ: Lawrence Erlbaum Associates, Inc.

Coltrane, S. (1998). Gender and families (1st ed.). Thousand Oaks, CA: Pine Forge. Cornell, D., Goldberg, J. P., & Folta, S. C. (2001). An intervention to increase fruit and vegetable con-

sumption using audio communications: In-store public service announcements and audio tapes. Journal of Health Communication, 6, 31–43.

Commonwealth of the Northern Mariana Islands, Food and Nutritional Advisory Council. (1996). CNMI nutrition policy and plan of action. Saipan CNMI: CNMI Government Printing Office.

Cox, M. J., & Paley, B. (1997). Families as systems. Annual Review of Psychology, 48, 243–267. Craig, R. T. (1999). Communication theory as a field. Communication Theory, 9, 119–161. Cronen, V. E., Pearce, W. B., & Harris, L. M. (1982). The coordinated management of meaning: A the-

ory of communication. In F. E. X. Dance (Ed.), Human communication theory: Comparative essays (pp. 61–89). New York: Harper & Row.

Cronk, L. (1999). That complex whole culture and the evolution of human behavior (1st ed.). Boulder, CO: Westview.

Dwyer, J. T., Stone, E. J., Yang, M., Webber, L. S., Must, A., Feldman, H. A., et al. (2000). Prevalence of marked overweight and obesity in a multiethnic pediatric population: Findings from the child and adolescent trial for cardiovascular health (CATCH) study. Journal of American Dietetic Association, 100, 1149–1154.

Ford, L. A., & Yep, G. A. (2003). Working along the margins: Developing community-based strategies for communicating about health with marginalized groups. In. T. Thompson, A. Dorsey, K. Miller, & R. Parrott (Eds.), Handbook of health communication (pp. 241–262). Mahwah, NJ: Lawrence Erlbaum Associates, Inc.

CHILDHOOD OBESITY 173

Frideres, J., & Goldenberg, S. (1982). Ethnic identity: Myth and reality in western Canada. Interna- tional Journal of Intercultural Relations, 6, 137–151.

Gabel, S., & Lutz, S. (2000). Household, parent, and child contributions to childhood obesity. Family Relations, 49, 293–300.

Goldenberg, I., & Goldenberg, H. (1999). Family therapy: An overview. Belmont, CA: Wadsworth. Goran, M. I. (2001). Metabolic precursors and effects of obesity in children: A decade of progress

1990–1999. American Journal of Clinical Nutrition, 73, 158–171. Heath, R. L., & Bryant, J. (2000). Human communication theory and research: Concepts, contexts, and

challenges. Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Kava, R., Meister, K. A., Whelan, E. M., Lukachko, A. M., & Mirabile, C. (2002). Dietary supplement safety

information in magazines popular among older readers. Journal of Health Communication, 7, 13–23. Klesges, R. C., Klesges, L. M., Eck, L. H., & Shelton, M. (1995). A longitudinal analysis of accelerated

weight gain in preschool children. Pediatrics, 95(1), 126–130. Laroche, M., Kim, C., & Tomiuk, M. A. (1999). Italian ethnic identity and its relative impact on the

consumption of convenience and traditional foods. British Food Journal, 101, 201–228. Larson, M. S. (1991). Health-related messages embedded in prime-time television entertainment.

Health Communication, 3, 175–184. McGoldrick, M., Giordano, J., & Pearce, J. K. (1996). Ethnicity and family therapy. New York:

Guilford. Meigs, A. (1997). Food as a cultural construction. In C. Counihan & P. Van Esterik (Eds.), Food and

culture: A reader (pp. 95–106). New York: Routledge. Mishler, E. G. (1984). The discourse of medicine: Dialectics of medical interviews. Norwood, NJ:

Ablex. Orbe, M. (2000). Centralizing diverse racial/ethnic voices in scholarly research: The value of

phenomenological inquiry. International Journal of Intercultural Relations, 24, 603–621. Papero, D. V. (1990). Bowen family systems theory. Boston: Allyn & Bacon. Pearce, K. A., & Pearce, W. B. (2001). The public dialogue consortium’s school-wide dialogue process:

A communication approach to develop citizenship skills and enhance school climate. Communica- tion Theory, 11, 105–123.

Pearce, W. B. (1976). The coordinated management of meaning: A rules-based theory of interpersonal communication. In G. Miller (Ed.), Explorations in interpersonal communication (pp. 17–36). Beverly Hills, CA: Sage.

Pearce, W. B., & Cronen, V. E. (1980). Communication, action, and meaning: The creation of social re- alities. New York: Praeger.

Pearce, W. B., & Pearce, K. A. (2000). Extending the theory of the coordinated management of mean- ing (CMM) through a community dialogue process. Communication Theory, 10, 405–423.

Philipsen, G. (1995). The coordinated management of meaning theory of Pearce, Cronen, and associ- ates. In D. Cushman & B. Kovacic (Eds.), Watershed research traditions in human communication theory (pp. 2–43). Albany: State University of New York Press.

Powdermaker, H. (1997). An anthropological approach to the problem of obesity. In C. Counihan & P. Van Esterik (Eds.), Food and culture: A reader (pp. 203–210). New York: Routledge.

Rasanen, M., Lehtinen, J-C, Niinikoski, H., Keskinen, S., Ruottinen, S., Salminen, M., Ronnemaa, T., Viikari, J., & Simell, O. (2002). Dietary patterns and nutrient intakes of 7-year old children taking part in an arteriosclerosis prevention project in Finland. Journal of the American Dietetic Associa- tion, 102, 518–524.

Reidy, C. A., Weinstein, P., Milgrom, P., & Bruss, M. (2001). An ethnographic study for understand- ing children’s oral health in a multicultural community. International Dental Journal, 51, 305–312.

Rimal, R. N., & Flora, J. A. (1998). Bidirectional familial influences in dietary behavior: Test of a model of campaign influences. Human Communication Research, 24, 610–637.

174 BRUSS, MORRIS, DANNISON, ORBE

Signorielli, N., & Lears, M. (1992). Television and children’s conceptions of nutrition: Unhealthy mes- sages. Health Communication, 4, 245–257.

Signorielli, N., & Staples, J. (1997). Television and children’s conceptions of nutrition. Health Commu- nication, 9, 289–301.

Socha, T. J., & Diggs, R. C. (Eds.). (1999). Communication, race, and family: Exploring communica- tion in black, white, and biracial families. Mahwah, NJ: Lawrence Erlbaum Associates, Inc.

Tardy, R. W., & Hale, C. L. (1998a). Bonding and cracking: The role of informal, interpersonal net- works in health care decision making. Health Communication, 10, 151–173.

Tardy, R. W., & Hale, C. L. (1998b). Getting “plugged in”: A network analysis of health-information seeking among “stay-at-home moms.” Communication Monographs, 65, 336–357.

U. S. Department of Commerce. (2000). Population and housing profile: 2000. the Commonwealth of the Northern Mariana Islands. Retrieved July 11, 2005, from http://www.census.gov/prod/cen2000/ island/CNMIprofile.pdf

Wolf, A. M., & Colditz, G. A. (1998). Current estimates of the economic cost of obesity in the United States. Obesity Research, 6(2), 97–106.

Wood, J. T. (1997). Communication theories in action. Belmont, CA: Wadsworth.

CHILDHOOD OBESITY 175