Examination of Weight Status and Dietary Behaviors of Middle School Students in Kentucky

Current Research

Examination of Weight Status and Dietary Behaviors of Middle School Students in Kentucky

MARY G. ROSEMAN, PhD, RD; WING KA YEUNG, MS, RD; JEN NICKELSEN, MS, RD ---- J Am Diet Assoc. 2007;107:1139-1145.

[1]ABSTRACT

Objective There is growing interest in understanding the roles that fruit, vegetable, milk, breakfast, and soft drink consumption play in relation to adolescents’ weight status. This study examines the relationship between weight status and dietary practices of middle school students.

Design This study consists of secondary analysis of a cross sectional survey given to 4,049 middle school students in central Kentucky.

Methods Students’ self-reported height and weight data, as well as 7-day recall of fruits, vegetables, milk, soft drinks, and breakfast consumption prior to completion of the survey, were collected. Self-reported height and weight were converted to body mass index (BMI) percentile according to Centers for Disease Control and Prevention’s classification criteria.

Outcome measures Measures included dietary consumption of fruits, vegetables, milk, soft drinks, and breakfast in relation to students’ BMI percentile.

Statistical analyses performed Descriptive statistics were extracted on demographics, BMI percentile, and food consumption. Bivariate analyses included x2 tests of association and Spearman rank correlation.

Results Healthy weight was associated with consuming fruits, vegetables, breakfast, and milk. Underweight and healthy-weight students consumed more fruits than stu- dents who were at risk of being overweight and over- weight. Healthy-weight students consumed more “other vegetables” than students who were at risk of being over- weight and more “other vegetables” and carrots than overweight students. Underweight students consumed breakfast more often than all other students, and healthy-weight students consumed breakfast significantly more frequently than students at risk of being overweight and overweight. Finally, overweight students had a significantly lower consumption of milk than all other students.

Conclusions Parents and school foodservice personnel should take these factors into consideration by developing menu strategies and tactics to encourage more healthful eating behaviors in children.

Children and teens who are overweight or at risk of becoming overweight have become a major health care concern in the United States. Results from the 1999-2002 National Health and Nutrition Examination Survey indicated that about 16% (over 9 million) of children and adolescents ages 6 to 19 years are overweight (1,2). Moreover, this proportion has tripled since 1980 (1,2). Kentucky youth are among the most overweight in the nation. According to the 2004 Youth Risk Behavior Survey, 17% of Kentucky middle school students were overweight, and another 20% were at risk for becoming overweight (3).

Overweight children are more likely to suffer from many chronic diseases, such as type 2 diabetes, high cholesterol, high blood pressure, early maturation, and orthopedic problems (4-7). In addition, they may have psychological problems and lower self-esteem (4-6,8). Overweight children tend to become obese adults (6,8), and being overweight is an important risk factor for development of chronic diseases later in life, including diabetes, stroke, arthritis, heart disease, and various types of cancer (2,7,8-10). Furthermore, the economic cost of health care for overweight children is increasing. For youth 6 to 17 years of age, annual hospital costs associated with being overweight increased more than threefold, from $35 million (1979 to 1981) to $127 million (1997 to 1999) (11).

Specific dietary factors related to children being overweight include: (a) consumption of high-energy-dense foods (10,12-15), (b) high dietary fat content and increased fast-food consumption (15-17), (c) high consumption of sugar-sweetened beverages and high-sugar foods (12,14,18-20), (d) low fruit and vegetable consumption (12,14,18), (e) large portion sizes (10,12,15), and (f) low breakfast consumption (21,22). Weight problems in children have been discussed in relation to environmental issues, such as food availability (23), family eating patterns (24,25), and socioeconomic status (24). Children eating dinner with their families at home have a better quality diet than those who do not (25). Family eating patterns, such as frequency of family meals, were positively associated with the intake of fruits, vegetables, grains, and calcium-rich foods and negatively associated with soft drink consumption (15,24,25). In the past few decades, there has been an increase in the consumption of soft drinks by children (15). At the same time, consumption of milk and vegetables has decreased (26).

Some of the reasons suggested for these dietary changes are the growing popularity and availability of fast foods and frozen dinners, as well as the mass marketing of unhealthful food to young children (10,15,16). Only 6% to 20% of children and adolescents nationwide ate five or more servings of fruits and vegetables per day (27-29). Another national study reported about two thirds of youth ages 12 to 19 consumed less than one serving of fruit a day and almost one third consumed less than one serving of vegetables a day (30). In addition, a nationally representative study of children ages 2 to 18 found soft drink, sweetened beverage, and fruit drink consumption had increased while milk consumption had decreased (19). Finally, children who ate breakfast had a better overall diet than those who did not eat breakfast (31-33) and an inverse relationship was found between body mass index (BMI) and breakfast consumption (21,22).

A number of previous studies investigating childhood weight factors have focused on elementary school children. However, elementary school children have less opportunity to make their own food decisions than middle school children who are evolving from set menus in primary school to cafeteria menu choices in middle school. Therefore, this study investigated the relationship be- tween weight status and dietary practices of middle school students in central Kentucky because it is a prime time for developing food-conscious choices and patterns for adult life. Specifically this study examined the relationship between weight status and consumption of fruits, vegetables, milk, soft drinks, and breakfast.

METHODS

Sample and Study Design

The current study analyzed cross-sectional data from a survey of middle school students in a central Kentucky public school district conducted in fall 2004. The county school system was comprised of 12 middle schools. The survey instrument used questions from the middle school Youth Risk Behavior Survey designed by Centers for Disease Control and Prevention to monitor the prevalence of priority health risk behaviors among youth (34). The Youth Risk Behavior Survey questionnaire has been found to be both reliable and valid for students in grades 7 to 12 (35-37). The middle school Youth Risk Behavior Survey was modified from the high school version to adjust for reading and comprehension abilities of students in that age group (38). A central Kentucky child- hood obesity prevention coalition added eight questions to the Youth Risk Behavior Survey as part of their efforts to assess local needs and evaluate interventions to combat childhood obesity. These questions were pretested with youth 9 to 13 years of age in the target population.

The coalition asked all public middle schools in the district to administer the survey to their students. Seven schools administered the survey to all students (sixth to eighth graders) and an additional school administered it to their eighth-grade students for a total survey population of 4,955. In most instances, the survey was administered during a single class period (usually the second). Passive parental consent was obtained via letters sent home to parents asking them to call the school if they did not wish their child to participate. The University of South Florida’s Institutional Review Board approved the protocol for secondary analysis of the dataset.

To meet the objectives of the study, in addition to the demographic variables, eight questions from the dietary behavior category were extracted regarding students’ self-reported dietary intake of fruits, vegetables, milk, breakfast, and soft drinks during the 7 days preceding the completion of the survey. Height and weight parameters were defined with reference to the age- and sex- specific growth charts by the Centers for Disease Control and Prevention (39). The present study applied the Centers for Disease Control and Prevention’s classification criteria: children with a BMI-for-age below the 5th per- centile were classified as underweight; children who were 285th percentile but 95th percentile were classified as at risk of being overweight; and children who were 295th percentile were classified as overweight (39,40). Self-re- ported height and weight were converted to BMI percen- tiles. The continuous variable for BMI was transformed to a categorical variable indicating whether the individual was underweight, healthy weight, at risk of being overweight, or overweight.

VARIABLE DEFINITIONS

In the survey, respondents were asked to self-report how often they consumed fruit, fruit juice, green salad, car- rots, “other vegetables,” milk, soft drinks, and breakfast during the past 7 days. Options for the fruit and vegetable questions were: none, 1 to 3 times per 7 days, 4 to 6

times per 7 days, or 1 time, 2 times, 3 times, or 4 or more times per day. For milk consumption, the options were the same except the term glass was used instead of “time.” Similarly, participants were asked to provide the number of times each day they drank a soft drink at school, home, or elsewhere with the options being: none, 1, 2, 3, 4, or 5 or more times per day. For breakfast intake, participants were asked how many days they ate break- fast during the past 7 days with the eight options being “none” to “all 7 days.”

Total consumption of each variable except breakfast was collapsed into fewer categories: did not consume in the past 7 days, 1 to 3 times in the past 7 days, 4 to 6

times in the past 7 days, 1 to 2 times per day, and 3 to 4 times per day. For breakfast, the collapsed categories were: did not consume, 1 to 3 days in the past week, 4 to 6 days in the past week, and every day.

Statistical Analyses

All analyses were conducted using SPSS (version 13.0, 2004, SPSS Inc, Chicago, IL). Descriptive statistics on demographics, body weight status, along with fruit, vegetable, milk, soft drink, and breakfast consumption were computed. Cross-classification tables and x2 tests were used to determine if the dietary consumption of fruit, fruit juice, green salad, carrots, “other vegetables,” milk, soft drinks, and breakfast was different by students’ weight status. For cross-classifications with significant associations, a post hoc pairwise analysis comparing each pair of weight categories for each food type and breakfast was conducted using the Pearson x2 test. Spearman rank correlation was conducted to determine the direction and strength of the association between weight status and dietary variables.

RESULTS

Description of Study Population

A total of 4,049 (81.7%) students with passive parental consent participated in the survey. Reasons for nonparticipation varied: lack of parental consent, absence from school on the day of the survey, and teachers who forgot to administer the survey. Missing responses in the data were a result of some students not answering a particular question. Ninety-eight percent of the sample population was 11 to 14 years of age, with about 49% female and 51% male. More than half of the participants described them- selves as white (61%), while one fourth described them as black or African American (25%). Other groups were Asian (5%), Hispanic or Latino (5%), American Indian or Alaska Native (2%), and Native Hawaiian or other Pacific Islander (2%). From the self-reported data on height and weight, only a small proportion of the participants were underweight (5%), more than half (64%) were at a healthy weight, approximately 16% (15% girls and 18% boys) were at risk of being overweight, and 15% (14% girls and 17% boys) were overweight.

Consumption of Fruits, Vegetables, Soft Drinks, and Breakfast Table 1 shows that during the past 7 days, about 25% of participants consumed fruit juice one time or more daily,

and about 30% consumed fruit one time or more daily.

Consumption of vegetables was much lower. About half of participants had not consumed a green salad or carrots during the past week, while 22% had not consumed “other

vegetables.” Only one third of participants consumed two or three glasses of milk daily. Examination of soft drink intake revealed that 29% of respondents consumed a soft drink once each day during the past week. Eighteen percent consumed soft drinks two times per day and 24% consumed soft drinks three times or more each day in the past week. For breakfast intake, slightly less than half of participants (45%) consumed breakfast every day, while 13% of participants did not eat breakfast during the past week. Twenty-two percent of respondents ate breakfast 1 to 3 days and 21% ate breakfast 4 to 6 days during the past week.

Weight Status and Fruit, Vegetable, Soft Drink, and Breakfast Consumption

No significant associations were found between students’ weight status and fruit juice, green salad, or soft drink consumption; therefore, these data are not provided. Findings indicate that weight status was significantly associated with consumption of breakfast (x2=53.94, P=0.01), milk (x2=31.67, P=0.002), “other vegetables” (x2=33.92, P0.001), and fruits (x2=53.94, P=0.01) (see Table 2). The association between weight status and car- rot intake approached statistical significance (x2=53.94, P=0.052). Results of a Spearman rank correlation analysis revealed that all associations were significant and in the negative direction (as weight increased, the frequency of food consumption decreased). However, all of these associations were weak (p<-0.12, P<0.002).

Table 3 provides post hoc x2 pairwise comparisons of BMI categories by food consumption variables. The significance of the x2 test implies that the frequency distribution of the particular food variable differed between the two paired weight categories. One can infer from Table 2 how intake differed between the two paired weight categories. Underweight students consumed breakfast more often than healthy-weight students, students who were at risk of being overweight, and overweight students. Healthy-weight students consumed breakfast more often than students who were at risk of being overweight and overweight. Underweight students, healthy-weight students, and students who were at risk of being overweight

all had higher milk consumption than overweight students. Healthy-weight students consumed more carrots than overweight students and more “other vegetables” than students who were at risk of being overweight or overweight. Both underweight and healthy-weight students consumed more fruit than students who were at risk of being overweight and overweight.