Volume 45, Issue 3, Supplement , Pages S30-S37, September 2009
Impact of School District Sugar-Sweetened Beverage Policies on Student Beverage Exposure and Consumption in Middle Schools
Article Outline
Abstract
Purpose
To determine the associations between 1) exposure to sugar-sweetened beverages (SSB) in middle schools and student consumption of SSB during the school day; and 2) school district policies about SSB and exposure to SSB in schools.
Methods
The strength of school district SSB policies was scored on three SSB policy indicators. Student SSB consumption at school was assessed by a self-administered Beverage and Snack Questionnaire. Exposure to SSB at school was defined as the number of vending slots and SSB venues as determined on-site at each school. Multivariate analysis considered the multilevel nature of the data.
Results
Data from 9151 students in 64 middle schools in 28 districts were used in the analysis. With schools as the unit of analysis, the proportion of students who consumed any SSB at school ranged from 19.2% to 79.8%. SSB exposure was a significant predictor of SSB consumption (β
=.157, p < .001). SSB consumption was not significantly associated with the size of the school, the racial or ethnic composition of the school's students, or the proportion of students eligible for free and reduced price meals. District SSB policy scores ranged from 0 to 6 with a mean score of 3.25 (±2.15). District SSB policy was a significant predictor of SSB exposure (β=−9.50, p < .0002).
Conclusions
School district SSB policies and exposure to SSB in middle schools are associated with student SSB consumption. Interventions to improve policies and their implementation may offer opportunities to improve the diets of adolescents.
Keywords: Nutrition policy, School, Adolescence, Sugar-sweetened beverages, Soft drinks
Most adolescents in the United States (US) drink sugar-sweetened beverages (SSB) on a daily basis. Data from the National Health and Nutrition Examination Study (NHANES) 1999–2004, indicate that 84% of youth aged 12–19 years consumed SSB on the surveyed day [1]. These beverages account for a substantial portion of the energy intake of youth in the US. Among youth who consumed at least one SSB on the survey day, consumption averaged 30 oz per day for a total of 356kcal. Although in the 1999–2004 data total energy from SSB did not differ by race or ethnicity, SSB consumption has increased disproportionately in youth from population groups at high risk for obesity. Compared with NHANES data from 1988–1994 there was a significant increase in per-capita SSB consumption among black and Mexican-American youth, but no significant change in per-capita consumption among white youth in the 1999–2004 data [1].
Most SSB provide energy but few nutrients. Adolescents who consume SSB often have diets that fail to meet recommended levels of nutrients such as calcium and vitamin D [2], [3], [4]. Data from the Continuing Survey of Food Intakes by Individuals indicate that only those children who are nonconsumers of SSB meet recommendations for calcium intakes [3]. Several recent research reviews that include experimental studies, especially those not funded by the beverage industry, conclude that there is an association between SSB consumption and obesity in children and adolescents [2], [5], [6], [7].
Published reports indicate that most adolescents in the US have ready access to SSB at school [8], [9], [10], [11]. In 2005, 80% of middle school students had access to vending machines with low-nutrient, energy-dense foods and beverages [11].
Several national organizations recommend limiting access to SSB in schools [4], [12], [13]. The Institute of Medicine Committee on Nutrition Standards for Foods in Schools [12] concludes that “in general, reducing consumption of sweetened beverages is viewed as an important component of a broad strategy to reduce excess energy intake.” The committee proposes standards that exclude carbonated, caffeinated, and most sweetened beverages in schools. The committee recommends that middle schools offer only water without flavoring, additives, or carbonation; 1% and nonfat milk, including flavored milks with <22g of total sugar per 8 oz and lactose-free and soy beverages; and 100% fruit juice in 4-oz portions.
Recently many states and school districts have adopted policies that limit access to SSB [14]. The Child Nutrition and WIC Reauthorization Act of 2004 required that every US school district that participates in United States Department of Agriculture (USDA) school meal programs establish wellness policies by the fall of 2006 [15]. The legislation indicates that these policies should include nutrition guidelines for all foods and beverages available at school. Across the US, there is considerable variation in the kinds of SSB policies that have been developed and the way they are being implemented. Among the 100 largest school districts in the US, 19% of districts have no mandatory nutrition guidelines for vending machines [16]. National data from the School Health Profiles survey indicate that in 2004 the median proportion of secondary schools across the states that allowed students to purchase soft drinks, sports drinks, or fruit drinks was 95.4% (range 78.9–99.5) [9]. By 2006, the median proportion of schools that allowed soda or sugar-sweetened fruit drinks was 62.5% (range 25.3–86.0), and the median that allowed sports drinks was 72.7% (range 30.5–90.2) [17]. In the 2006 School Health Policies and Programs survey, 80.4% of all districts still allowed schools to sell soda, sports drinks, or fruit drinks [18].
School SSB policies have the potential to address important adolescent health problems, but studies of these policies tend to be few and limited in scope. For adolescents, frequency of soft drink consumption is positively associated with the use of vending machines and access to soft drinks at school [19], [20], and a study in three Texas middle schools found that implementation of state standards reduced student SSB consumption during lunch from 5.4 to 1.5 oz. [21]. Important concerns have been raised about the potential for inequities in local school policies to increase health disparities among youth if they result in improved school environments only in schools that serve affluent students [22].
In the state of Washington, there are no state laws or regulations that set specific requirements for the nutritional content of foods and beverages in schools, and each school district has developed separate nutrition policies. The purpose of this study is to take advantage of this “natural experiment” to measure the associations between 1) exposure to SSB in middle schools and student consumption of SSB at school; and 2) school district SSB policies and exposure to SSB in schools.
Methods
Recruitment
School recruitment was designed to ensure representation from middle schools that serve low-income and racial/ethnic minority students and geographically diverse parts of Washington State by using existing relationships to engage schools. The project's partners, the Washington State School Directors Association, Office of the Superintendent of Public Instruction and public health advocates, announced the project at meetings, through newsletters, in letters mailed directly to school administrators and leaders and through personal contact. All public schools that enroll seventh-grade students and participate in USDA school meal programs were eligible to participate. A total of 65 schools from 29 districts returned a letter of cooperation agreeing to participate in all phases of data collection. Research tools and protocol were approved by the University of Washington Institutional Review Board.
SSB policy score
School district policies were obtained from the Office of the Superintendent of Public Instruction. Representatives from each of the districts were asked to verify that these were the most up-to-date versions of the policies at the time that data were collected in the schools. The strength of school district policies was scored by research staff using a three item SSB subset from a 96-item coding system that was developed and tested by a national team of nutrition policy researchers [23]. The overall interrater reliability score for individual items on the coding system is .72. The three SSB policy quality indicators are as follows: limits sugar content of beverages, limits regular (sugar-sweetened) soda, and limits beverages other than soda containing added caloric sweeteners such as sweetened teas, juice drinks, energy drinks and sports drinks. If the policy does not mention a topic at all, a score of 0 is assigned. If the topic is addressed but stated in weak or suggested language (for example, “only beverages that are low in sugar will be available at school”), a score of 1 is assigned. If a topic is addressed in a specific and measurable way (for example, “only beverages that have less than 10g of added sugar will be available at school”), a score of 2 is assigned. The scores for the three items were summed for each district to obtain a SSB policy score.
Exposure to SSB at school
During the 2007–2008 school year, beverages available for purchase by students outside USDA school meal programs were recorded on-site at each school. Data were entered into FoodBEAMS, a school nutrition environment assessment tool [24]. Each type and brand of an item available was recorded; in vending machines each slot was recorded. The FoodBEAMS database was expanded to include nutrient composition information for each beverage. Beverages available for purchase in each school were categorized as either containing added sugars or not containing added sugars to test for alignment with the recommendations of the Committee on Nutrition Standards for Foods in Schools [12]. Dairy products with added sugars were not included in measures of SSB exposure because of the committee's recommended exemption of some of some dairy products based on the importance of calcium-rich beverages [12]. Using a modified version of the methods of Rideout et al [25], an index of SSB availability was calculated for each school as a measure of exposure to SSB at school. This index was calculated as the [(no. of SSB exposures)/ (no. of students enrolled in the school)]×1000. SSB exposures were defined as either one vending machine slot, or the availability of each unique SSB in the school cafeteria, snack bar/cart, or school store.
Student SSB consumption
During the 2007–2008 school year, the research team worked with participating schools to administer the Beverage and Snack Questionnaire to all seventh graders [26]. To maintain student anonymity, the research team provided the schools with parent letters, student assent forms, a teacher script, and the survey tool. The schools provided the letters to families asking them to return signed forms if they did not want their student to participate. Student assent was obtained. Each school decided how they would administer the surveys, either all at once (for example during first period) or during a specific class (for example, during homeroom). The schools returned the completed surveys to the research team. The Beverage and Snack Questionnaire has been found to have adequate validity (r=.72–.85) and reliability (r=.69–.71) with seventh graders [26].
School characteristics
District and school level demographic data were obtained from the Washington State Office of Superintendent of Public Instruction [27]. Data about the school's geographic locale were obtained from the National Center for Education Statistics [28].
Statistical analysis
The unit of analysis was schools. School-level data were combined into a database that described school characteristics, district policies, exposure to SSB, and proportions of students who consumed SSB during the school day. SPSS (version 16.0; SPSS Inc., Chicago, IL) and SAS (version 9.1; SAS Institute, Cary, NC) were used to analyze descriptive data and to test for associations. Pearson's correlation coefficients were used to examine relationships between continuous variables. Analysis of variance with Bonferroni comparisons and t tests were used to examine differences between groups. SAS Proc Mixed was used for the multivariate analysis because of the multilevel nature of the data: schools (level 1) nested in districts (level 2) [29]. The analysis used a random intercept model (random effect of district), which corrects for the nonindependence of observations that would lead to lower standard errors and increased type II errors.
Results
Characteristics of schools
Complete data, representing 28 school districts, were collected in 64 of the 65 schools that volunteered to participate in the study. One school (and consequently one district) withdrew from the study because of widespread flooding and school closures. In 2008 there were 77,731 seventh-grade students in 523 schools in 295 districts in Washington State. Questionnaires were collected from 10,618 students. The total seventh-grade enrolment in the 64 study schools was 13,889, for a total response rate of 76%. The mean absenteeism rate among schools that publicly report these rates was approximately 10%. The schools and the exposure to beverages at each school are described in Table 1. The demographic characteristics of the students in the study schools were similar to those in the state as a whole. In Washington State, the mean proportion of students who were eligible for free and reduced-price school meals (i.e., free and reduced-price eligible) was 38%; in the study schools 38.9% were eligible for the programs. Overall, in Washington State the proportion of students in each racial/ethnic category was 8.7% Asian/Pacific Islander, 2.7% American Indian/Alaska Native, 5.7% black, 14.6% Hispanic, and 65.8% white. In the study sample the proportion of students in each racial/ethnic category was 5.8% Asian/Pacific Islander, 3.0% American Indian/Alaska Native, 4.3% black, 11.2% Hispanic, and 75.0% white.
Table 1. Characteristics of schools studied and exposure to beverages at school (n = 64)
| Characteristic | Total | Mean (SD) | Range |
|---|---|---|---|
| School characteristics | |||
| School enrollment | 630 (233) | 95–1074 | |
| Size of district | |||
| Schools in large districts (>9000 students) | 40 | ||
| Schools in medium districts (2000–9000 students) | 18 | ||
| Schools in small districts (<2000 students) | 6 | ||
| Proportion of students free and reduced-price–eligible | 38.9% (18.6) | 2.6–93.3% | |
| schools with >30% FRPM | 42 | ||
| Race/ethnicity | |||
| Asian / Pacific Islander | 5.8% (6.9) | 0.1–37% | |
| American Indian/ Alaskan Native | 3.0% (4.1) | 0.2–25% | |
| Black | 4.3% (6.5) | 0.6–46% | |
| Hispanic | 11.2% (13.1) | 1.5–82% | |
| White | 75.0% (18.4) | 3.6–94% | |
| Schools with >30% nonwhite | 19 | ||
| Schools with >30% Hispanic | 4 | ||
| Geographic distribution | |||
| City | 24 | ||
| Noncity | 40 | ||
| Exposure to nondairy beverages at school: Total Number of exposure slots∗ | |||
| Sugar-sweetened beverages | |||
| Sodas | 2.0 (4.9) | 0–19 | |
| Other beverages with sugar∗∗ | 12.0 (12.6) | 0–53 | |
| Beverages that are not sugar sweetened | |||
| Water | 11.3 (14.6) | 0–80 | |
| 100% Juice | 10.4 (11.0) | 0–66 | |
| Diet soda | 0.6 (1.9) | 0–10 | |
| Artificially sweetened non-soda beverages | 5.9 (7.2) | 0–47 | |
| Exposure to beverages at school by enrollment | |||
| No. of total exposure slots/ total enrollment×1000 | 77.5 (66.7) | 4–347 | |
| No. of total exposure slots for SSB/Total enrollment×1000 | 28.7 (35.0) | 0–147 |
∗An exposure slot is defined as one slot on a vending machine or the availability of a beverage in the cafeteria (other than those provided as part of the school lunch or breakfast program), snack bar, or student store. |
∗∗“Other beverages with sugar” are defined as non-soda beverages that contain any grams of added sugar, including sweetened juice drink, carbonated sweetened juice drink, sports drink, sweetened coffee, sweetened tea drink, hot chocolate, and sweetened water drink. |
School district SSB policies
Table 2 provides information about the distribution of district scores on the three indicators of SSB policy strength. Additive SSB district policy scores ranged from 0 to 6, with a mean of 3.25 (±2.15).
Table 2. Categorization of sugar-sweetened beverage policy indicators (n = 28)
| Not mentioned (score=0) | Stated in weak or suggested language (score=1) | Stated in a specific and measurable way (score=2) | |
|---|---|---|---|
| Degree to which policy: | |||
| Limits sugar content of beverages | 13 | 4 | 11 |
| Limits availability of soda | 5 | 10 | 13 |
| Limits beverages other than soda containing added caloric sweeteners | 8 | 14 | 6 |
Access to SSB in schools
Three of the schools did not sell beverages outside of the cafeteria (Table 3). A total of 61 schools had beverage vending machines. In 33 schools, nondairy beverages were sold in the cafeteria. In two schools, beverages were sold from a snack cart, and in nine they were sold in a student store or snack bar. In 17 schools no SSB were sold on campus.
Table 3. Access to beverages in schools (n = 64)
| Beverage category | Available on campus: no. of Schools | Available in vending machines: no. of Schools | Available in the cafeteria, snack bar/cart, school store: no. of schools | Vending machines per school: mean (SD) | Vending slots per school: mean (SD) |
|---|---|---|---|---|---|
| Sugar-sweetened beverages | |||||
| Soda | 11 | 11 | 1 | 1.16 (2.9) | 2.00 (4.8) |
| Other beverages with sugar∗ | 46 | 38 | 25 | 5.33 (7.0) | 9.22 (11.6) |
| Beverages not sugar sweetened | |||||
| Water | 62 | 58 | 32 | 1.30 (1.3) | 10.7 (14.6) |
| 100% juice | 60 | 47 | 51 | 2.88 (5.0) | 7.27 (10.6) |
| Diet soda | 13 | 13 | 0 | 0.55 (1.6) | 0.63 (1.9) |
| Artificially sweetened non-soda beverages | 46 | 38 | 13 | 2.67 (2.9) | 5.42 (7.2) |
∗“Other beverages with sugar” are defined as non-soda beverages that contain any grams of added sugar, including sweetened juice drink, carbonated sweetened juice drink, sports drink, sweetened coffee, sweetened tea drink, hot chocolate, and sweetened water drink. |
Student beverage consumption at school
The index of student SSB consumption at school was the proportion of students at each school who reported consuming any SSB during the school day. Student SSB consumption was calculated with individual data from five questions on the Beverage and Snack Questionnaire (Appendix 1). From the 64 schools, a total of 10,618 questionnaires were collected, and a total of 9,151 were used to calculate the student beverage consumption scores. Questionnaires were excluded for the following reasons: 1) missing two or more of the five items and 80% from the entire questionnaire (970; 9%); 2) missing 80% or more items from the entire questionnaire (291; 3%); 3) invalid responses such as filling in more than one answer for a question (137; 1.3%); 4) missing two or more of the five items (53; .5%); and 5) missing school information (16, .2%). For the 647 students who were missing data for only one of the five questions, the mean of the other four items was substituted for the missing item. Across schools, the proportion of students who consumed any SSB at school ranged from 19.2% to 79.8%, with a mean of 57.86% ± 12.88.
Variables associated with SSB consumption at school
The proportion of students who consumed SSB at each school was positively associated with SSB exposure at school (Pearson's correlation coefficient .40, p=.001). The percentage of students who consumed SSB at school was higher in city schools (mean 61.26% ± 12.41) compared with the noncity schools (mean 52.19% ± 11.34) (two-sided t statistic 2.88, df=62, p=.005), but was not significantly associated with the size of the school, the racial or ethnic composition of the school's students, or the proportion of students who were free and reduced-price eligible. The proportion of students who consumed SSB at school was inversely associated with the quality of SSB policies as measured by the overall sum of the district policy score (F statistic 4.25, df=6, p=.001). When schools that had no SSB policies (total SSB policy score 0) were compared with those that had some policies, the proportion of students who were consuming SSB at school was higher than those in schools that had more or stronger policies that addressed SSB. These comparisons of schools with scores of 0 were significant for those with scores of 3 (p=.034,) 4 (p=.004), and 6 (p=.041). There was not a statistically significant association between districts with a policy score of 0 compared with those with a score of 1, 2, or 5.
Multivariate analysis
The first analysis explored the question of whether the proportion of students drinking any SSB at a school is predicted by exposure to SSB. The dependent variable is the proportion of students drinking any SSB and the independent variable of interest was the index of SSB availability. With only 64 schools, a limited set of school level variables were included in the model. The three control variables that were thought, a priori, to be the most important were: 1) whether the school was located in a city, 2) the percentage of students eligible for free and reduced-price meals, and 3) the percentage of white students at the school. The first step was to verify that variability in SSB consumption existed between schools and between districts. This model showed marginally significant between-district variance in SSB consumption (40.81, p < .064) and quite significant between-school variance in SSB consumption (123.49, p < .0001). The second step was to run the model with the variables described above, entered into the model simultaneously. Using a random intercept model, none of these control variables were statistically significant (−4.38, p=.27; .04, p=.71; and −.14, p=.28, respectively), only the variable of interest, exposure to SSB, was statistically significant. Given that location (city, noncity) was significant in the bivariate analyses, this variable was tested as the only control. It was not statistically significant in this model either (−3.68, df=34, p=.308). Hence these three control variables were dropped from the final model leaving only exposure to SSB as the predictor of consumption. The final model showed that SSB exposure was a significant predictor of SSB behavior (β=.16, p = .001) in the expected direction: that is, more availability of SSB at a school leads to a higher percentage of students drinking SSB. This model with parameter estimates is:
The reduction in between-school variance (change in R2) in SSB consumption due to SSB exposure was 17%. This result is shown in Figure 1. The following hypothetical example gives a sense of the effect size estimated by this model for a school with 500 students. Changing the offerings in one typical nine-slot vending machine from SSB to non-SSB would reduce the percentage of students drinking any SSB at school by 2.83%.
Figure 1
Exposure at school (slots of sugar sweetened-beverage [SSB] /per student
×1000) versus percentage of students drinking any SSB at school with fitted line.
The second analysis explored the question of whether SSB policy at the district level, as measured by the total of three items as described above, was a significant predictor of the number of SSB slots at schools as measured by our modified Rideout formula. The unconditioned means model showed significant variance in SSB exposure at both the district level (1181.41, p = .002) and the school level (382.27, p < .0001). Results show that district SSB policy is a significant predictor of SSB exposure (β=−9.50, p = .004): that is, a stronger anti-SSB policy leads to less exposure. This model with estimated parameters is:
According to this model, SSB policy accounts for 28% of the variance at the district level in exposure. This result is shown in Figure 2.
Figure 2
School policy score versus exposure (total slots of sugar sweetened-beverage (SSB) per student
×1000) with fitted line. For each policy score, the same symbol indicates schools that belong to the same district (see text for additional discussion).
Our results show that policy influences exposure and that exposure influences consumption in the expected directions. Districts with better policies have schools with less SSB exposure, and schools with less SSB exposure have fewer students consuming SSB. Not surprisingly, policy also statistically significantly influences consumption in the expected direction when it is the lone predictor in a random intercept model (−2.43, p < .004). However, it does not have a statistically significant effect on consumption over and above that of exposure when both are in the model (−1.49, p=.09). Although not a direct test, this is consistent with the conceptual model that policy influences consumption via the mediating variable exposure.
Discussion
This study provides information about school SSB policies, student exposure to SSB at school and student consumption during the 2007–2008 school year, when the national 2006 school wellness policy requirement had been in effect for a year. The variation in school district nutrition policies, student demographics and geographic locales across the state of Washington provided an opportunity to explore the impact of polices in diverse settings. The major findings are that exposure to SSB in middle schools is associated with student SSB consumption at school and that school district policies are associated with students' exposure to SSB at school.
There are limitations to school nutrition policy research methods, including those used in this study. First, given that an assortment of beverages is usually available at several locations within schools, it is not clear what that best measurement of student exposure to SSB should be. Second, self-report of dietary behaviors is subject to bias and inaccuracy, especially in children. Third, there is the problem of unmeasured variables, in particular the sources of SSB that were consumed at school. Some of the SSBs consumed at school might have come from home or neighborhood stores, and this consumption would not be influenced by school policies or exposure. Fourth, the 64 schools were a volunteer sample of only about 10% of the school districts in Washington State. Hence it is unclear to what population the results generalize.
A strength of this study is that the school nutrition environment was measured directly; most nationally representative studies of school nutrition environments have relied on responses from local school administrators and little is known about their ability to provide valid and reliable answers.
Exposure to SSB at school and student SSB consumption
SSB are still common in middle schools, and exposure to SSB in schools is associated with student beverage choices. Nationally, in 2005 65% of eighth graders were reported to have access to regular soft drinks in vending machines at school, and 48% had access to SSB in the school cafeteria [10]. SSB were sold to students in 47 of 64 schools in this sample. This exposure is associated with SSB consumption in this and previous studies. Using adjusted logistic regression models that included personal taste, access to soda at home, and interpersonal variables, Grimm et al found that children aged 8–13 years were 2.4 times more likely to consume soft drinks more than five times a week when soda was available at school [19].
The determinants of student dietary behaviors are complex. In this study, SSB exposure accounted for 17% of the between-school variability in SSB consumption; however, that still leaves 83% of the variance unaccounted for. Using an adjusted regression model to predict SSB intake in middle school students that included both school vending machine use and visits to fast-food restaurants Wiecha et al [20] were able to predict just 10% of the variance in SSB consumption. Figure 1 shows much heterogeneity in consumption even at schools with zero exposure; clearly other factors are important. Although we did not find significant influences of geography, ethnicity, and income once exposure was in the model, this is may be caused by our crude measures (city/noncity, percent white, and percent free and reduced price eligible) and sample size.
Policies and SSB exposure
Despite the variability between schools within the same district, there was an overall impact of district policies on SSB exposure in schools. As shown in Figure 2, in one district (open triangles) with the highest possible policy score of 6, there was a wide range of SSB exposure scores, with values for individual schools of 11.5, 14.5, 41.7, 58.9, 62.8, and 82.9. In another school district (solid circles) with the lowest possible policy score of zero, six of the schools had very similar exposure scores with a mean of 13.8, but one school in this district had an exposure score of 65.4 (close to the value predicted by the model). The differences in SSB exposure in schools within districts, indicates the need for further investigation into factors such as the implementation, monitoring, and enforcement of district policies. A 2008 report on the status of school wellness policies states that, “Tracking of local wellness policy implementation is as important as development of the policy itself, yet monitoring and evaluation are not given the attention necessary, and in many districts are virtually nonexistent” [30].
School SSB policy and inequities
There is a potential for wellness policies to contribute to ongoing health disparities [22]. In this study, there was no statistically significant added influence of the demographic characteristics of the students on the associations between policy and exposure to SSB. Other researchers have reported similar results with larger samples. Nationally representative data collected from 395 schools in 2005 indicate that neither the proportions of students who are free and reduced price eligible nor the proportion of racial/ethnic minority students were associated with a food environment summary score [11], and the Youth, Education and Society study in 2005 found no significant socioeconomic differences in student SSB exposure at school [10]. Results from these studies indicate that there may even be a trend toward better policies and environments in schools with a higher proportion of racial/ethnic minority students. The process of school nutrition policy development and implementation is not well understood, but rigorous and effective school SSB policies may have the potential to alleviate health disparities associated with consumption of SSB.
It is noteworthy that a wide variety of sugar-sweetened beverages are now on the market. In this study SSB such as sweetened juice drink, carbonated sweetened juice drink, sports drink, sweetened coffee, sweetened tea drink, hot chocolate, and sweetened water drink were more common than sodas. Schools may be providing these SSB that have the same nutritional content as sodas in the belief that they are healthier. In a quasi-experimental study in Maine, when schools were provided with lists of beverages that met nutrition guidelines and received help in communicating with vending suppliers, the proportion of beverage vending rows that met the guidelines increased from 48% to 99% [31]. Similar informational interventions may be needed to address confusion about the nutritional qualities of SSB.
Findings from this study suggest that school SSB policies impact exposure to SSB at school, and that SSB exposure at school impacts student SSB consumption at school. Although nutrition policy can only explain part of the variability in student SSB consumption at school, it appears that school districts that have strong policies in place can be assured that this is a positive step toward promoting the health of their students.
Acknowledgments
The authors thank the Healthy Eating Research Work Group One including Marlene Schwartz, Anne Lund, Mollie Greves Grow, Leslie Lytle, Anne Samuelson, Claudia Probart and Elaine McDonnell for sharing their expertise in developing the school wellness policy coding tool and the data collection team including Mary O'Leary, Ashley Hardesty, Tennyson Salopek, Tino Alonso, April Davis and Julie Larsen. This research was funded by the Robert Wood Johnson Foundation's Healthy Eating Research program grant 57932 and Active Living Research program grant 61125.
Appendix 1
Beverage and Snack Questionnaire, Sugar Sweetened Beverage (SSB) questions:
How often did you drink these beverages in the past week?
Question 1: Fruit drinks (such as Snapple, flavored teas, Capri Sun, and Kool-Aid)
Question 2: Sport drinks (such as Gatorade or Powerade); these drinks usually do not have caffeine
Question 3: Flavored waters such as Propel or vitamin waters; these drinks usually do not have caffeine
Question 4: Regular soda or pop (including all kinds such as Coke, Pepsi, 7-Up, Sprite, root beer)
Question 5: Energy drinks (such as Rockstar, Red Bull, Monster, and Full Throttle); these drinks usually have caffeine
Answer choices: Never or less than one per week, one per week, two to four per week, five to six per week, one per day, two to three per day, four or more per day.
References
- . Increasing caloric contribution from sugar-sweetened beverages and 100% fruit juices among US children and adolescents, 1988–2004. Pediatrics. 2008;121:1604–1614
- . Effects of soft drink consumption on nutrition and health: A systematic review and meta-analysis. Am J Public Health. 2007;97:667–675
- . Children and adolescents' choices of foods and beverages high in added sugars are associated with intakes of key nutrients and food groups. J Adolesc Health. 2004;34:56–63
- . Soft drinks in schools. Pediatrics. 2004;113:152–154
- . The role of sugar-sweetened beverage consumption in adolescent obesity: A review of the literature. J Sch Nurs. 2008;24:3–12
- . Intake of sugar-sweetened beverages and weight gain: A systematic review. Am J Clin Nutr. 2006;84:274–288
- How discretionary can we be with sweetened beverages for children?. J Am Diet Assoc. 2008;108:1440–1444
- US Government Accountability Office. School meal programs: Competitive foods are widely available and generate substantial revenues for schools. US Government Accountability Office, 2005 Report No. GAO-05-563.
- Competitive foods and beverages available for purchase in secondary schools. MMRW Weekly. 2005;54:917–921
- . Soft drink availability, contracts, and revenues in American Secondary Schools. Am J Prev Med. 2007;33:S209–S225
- . School food environments and policies in US public schools. Pediatrics. 2008;122:e251–e259
- . Nutrition standards for foods in schools. In: Stallings VA, Yaktine AL editor. Leading the Way toward Healthier Youth. Washington, DC: National Academies Press; 2007;
- National Alliance for Nutrition Activity (NANA). Model School Wellness Policies [Online]. Available at: http://www.schoolwellnesspolicies.org. Accessed September 12, 2006.
- . The interplay of public health law and industry self-regulation: The case of sugar-sweetened beverage sales in schools. Am J Public Health. 2008;98:595–604
- Public Law 108-265, 119 Stat 729. Child Nutrition and WIC Reauthorization Act of 2004 [Online]. Available at: http://www.fns.usda.gov/cnd/Governance/Legislation/Historical/PL_108-265.pdf. Accessed June 15, 2008.
- . A Foundation for the Future: Analysis of Local Wellness Policies from the 100 Largest Districts. School Nutrition Association. 2006;
- . Competitive Foods and Beverages Available for Purchase in Secondary Schools—Selected Sites, United States, 2006. Atlanta: Georgia: Centers for Disease Control and Prevention; 2008;
- Nutrition services and foods and beverages available at school: Results from the School Health Policies and Programs Study 2006. J Sch Health. 2007;77:500–521
- . Factors associated with soft drink consumption in school-aged children. J Am Diet Assoc. 2004;104:1244–1249
- School vending machine use and fast-food restaurant use are associated with sugar-sweetened beverage intake in youth. J Am Diet Assoc. 2006;106:1624–1630
- . Improvements in middle school student dietary intake after implementation of the Texas public school nutrition policy. Am J Public Health. 2008;98:111–117
- . Evaluating the distribution of school wellness policies and practices: A framework to capture equity among schools serving the most weight-vulnerable children. J Am Diet Assoc. 2008;108:1436–1439
- Schwartz MB, Lund AE, Grow HM, et al. A comprehensive coding system to measure the quality of school wellness policies. J Am Diet Assoc (in press).
- Samuels & Associates. Food and Beverage Environmental Analysis and Monitoring System (Food Beams). Oakland, California; 2007.
- Food sales outlets, food availability, and the extent of nutrition policy implementation in schools in British Columbia. Can J Public Health. 2007;98:246–250
- . Development and validation of a beverage and snack questionnaire for use in evaluation of school nutrition policies. J Am Diet Assoc. 2009;(in press)
- Washington State Office of Superintendent of Public Instruction. Washington State Report Card, 2007 [Online]. Available at: http://reportcard.ospi.k12.wa.us/DataDownload.aspx. Accessed October 3, 2008.
- US Department of Education. Common Core of Data (CCD) 2005–2006 [Online]. Available at: http://nces.ed.gov/ccd. Accessed October 3, 2008.
- . Hierarchical Linear Models: Applications and Data Analysis Methods. 2nd ed. Thousand Oaks, California: Sage Publications; 2002;
- . What's Working for and against Healthy Schools. Online Report. Skokie, Illinois: Action for Healthy Kids, Fall. 2008;
- Implementation of low-fat, low-sugar, and portion-controlled nutrition guidelines in competitive food venues of Maine public high schools. J Sch Health. 2007;77:687–693
PII: S1054-139X(09)00118-9
doi:10.1016/j.jadohealth.2009.03.008
© 2009 Society for Adolescent Medicine. Published by Elsevier Inc. All rights reserved.
Volume 45, Issue 3, Supplement , Pages S30-S37, September 2009
