|Home | About | Journals | Submit | Contact Us | Français|
Compared to students attending regular high schools, alternative high schools (AHS) students are more likely to be racial/ethnic minorities, have higher levels of poverty, and higher rates of risky and poor health behaviors, including weight-related behaviors, such as limited fruit/vegetable intake. However, little is known about fruit/vegetable intake among AHS students. This study examined whether perceived barriers to healthy eating mediated association between self-efficacy to eat healthy foods and fruit/vegetable consumption among AHS students. The cross-sectional study population consisted of students (n=145) attending 6 alternative high schools in the St. Paul/Minneapolis, MN area who were participants in an obesity prevention pilot study and completed a baseline survey in Fall 2006. Mixed model linear regression, adjusting for sociodemographic characteristics was used to test a series of regression models performed according to mediation analysis procedures. Students' mean age was 17.3 years, 52% were male, 63% were low-income, and 61% were from racial/ethnic minorities. Students reported a mean fruit/vegetable intake of 3.6 servings per day, mean self-efficacy to eat healthy score of 22.2 (range 3-35), and mean barriers to eating healthy score of 6.9 (range 3-13). Perceived barriers to healthy eating fully mediated the relationship between self-efficacy and fruit/vegetable consumption (Sobel test statistic=2.7, p=0.007). Interventions targeting the dietary practices of AHS students should include components to decrease perceived barriers as a way to increase self-efficacy and ultimately fruit/vegetable intake.
Increasing fruit and vegetable consumption has been identified as a major public health priority (1, 2). Among adolescents, most do not consume the recommended nine servings of fruit/vegetables per day (2, 3). Studies indicate that some groups of youth may be at greater risk for lower fruit/vegetable intake. For example, older youth report consuming fewer fruits/vegetables than younger youth (3, 4). Studies also support a positive association between income and fruit/vegetable intake with low-income youth consuming fewer fruit/vegetables than their higher income counterparts (5, 6). The association with race/ethnicity is less consistent (3-4, 7-8).
Reaching low-income, minority adolescents with positive nutrition messages is important as these young people are more at risk for obesity and related chronic disease (1). However, gaining access to this population of young people can be challenging (9). Alternative high schools are educational settings for students at risk for drop-out from traditional schools due to problems such as truancy, teen pregnancy, and behavioral problems (10, 11). Alternative schools provide access to youth who are more likely to be older, racial/ethnic minorities, have higher levels of poverty, and higher rates of risky health behaviors that include less healthy dietary practices (10-16). Yet little is known about factors that might contribute to improving the dietary behaviors of students.
The purpose of the current study was to enhance understanding of factors associated with fruit/vegetable consumption among adolescents attending alternative high schools. Guided by Social Cognitive Theory (17, 18), it was hypothesized that perceived barriers to healthy eating (socio-environmental factor) would mediate the association between self-efficacy (personal factor) and fruit/vegetable consumption (behavior). Several studies have assessed adolescents' food choices (7, 19), correlates to food selections, including self-efficacy (20, 21), and barriers to healthy eating (22, 23). However, few studies have assessed possible pathways that may contribute to behavior change (24). Mediation models can provide a means to improve understanding of behavioral pathways (25, 26).
The current study was a cross-sectional secondary analysis of baseline data obtained from the Team COOL (Controlling Overweight and Obesity for Life) pilot study. This group randomized school-based nutrition and physical activity intervention was conducted with a convenience sample of four urban and two suburban alternative high schools in Minneapolis-St. Paul, Minnesota (16). Across the six schools, the average enrollment was 102 students (range: 27-142 students), mean age was 17 years (range: 14-20 years), 52% were male, 64% racial/ethnic minorities (range 31-96%), and 60.5% qualified for free/reduced school meal programs (range: 40-96%).
In Fall 2006, prior to randomization, all students enrolled in study schools were invited to complete a survey. Informed consent was obtained from participants 18 years and older, while those less than 18 years provided signed parental consent and individual assent. The self-administered 76-item survey was completed by 145 students and took approximately 30 to 40 minutes to complete. Details regarding survey measures have been previously published (16). Participants received a $5 gift card for completing the survey and height/weight measures. The University of Minnesota Institutional Review Board reviewed and approved the study.
Students attending alternative high schools have lower attendance rates than traditional school students (10-12). The study participation rate was therefore calculated by multiplying the school's 2006-2007 enrollment by the school's 2005-2006 attendance rate (15,16, 27). The adjusted participation rate across the six schools for the baseline survey was 36% (Range: 18%-100%).
Fruit/vegetable consumption was assessed using a previously validated six-item fruit and vegetable screener (28), which assesses past-year consumption: Students were asked “Think about your usual eating habits over the past year. About how often do you eat each of the following foods and beverages?: (a) 100% Juice, (b) fruit (not including juice), (c) green salad (d) potatoes (not including French fries, fried potatoes, and potato chips), (e) carrots, (f) vegetables (not including carrots, potatoes or salad).” Response categories ranged from less than once a week to five or more times per day. Similar to other studies, data were recoded as daily servings of fruits/vegetables and summed (15, 29). Cronbach's α for the study sample was 0.85. Responses were considered outliers and excluded if they were greater than three standard deviations above the mean (n=2) (15).
Perceived barriers to healthy eating was assessed with a three-item scale used previously with adolescents and found to be reliable (30). Students responded to the question “How strongly do you agree with the following statements: (a) I am too busy to eat healthy foods, (b) Healthy foods cost too much, and (c) Eating healthy just takes too much time. Response categories ranged from strongly agree to strongly disagree on a five-point Likert scale. Responses were summed, with higher score indicating more barriers. Cronbach's α for the study sample was 0.82.
Self-efficacy to eat healthy was assessed with a seven-item scale (31). Participants were asked the following question: “How confident are you that you could change or maintain your eating patterns to: (a) Eat at least two servings of fruit, (b) Eat at least three servings of vegetables (not including fried potatoes), (c) Limit the soda pop you drink, (d) Limit how often you eat at fast food restaurants, (e) Limit between meal snacks, (f) Limit the portion sizes of the food you eat, and (g) Eat breakfast on most days.” Response categories ranged from not at all confident to very confident on a five-point Likert scale and were summed, with a higher score indicating greater confidence. This question has been shown to be reliable in other studies with adolescents (31). Cronbach's α for the study sample was 0.84.
Gender and age were determined using school records. Race/ethnicity was measured with the item “Do you think of yourself as American Indian/Alaskan; Asian; Black or African American; Hispanic/Latino; White; Other.” Due to small samples within several of the racial/ethnic groups, the “Other” category was recoded to include the following groups: American Indian/Alaskan (1%); Asian (6%); Hispanic/Latino (9%); multiethnic non-Hispanic (10%); other (3%). Socioeconomic status was assessed with the question: “Do you get free/low-cost lunches at school?” Students who reported having received free/reduced lunch were categorized as low income. If students did not respond to this question (n=8), but responded yes to participating in federal nutrition assistance programs, they were categorized as low-income.
Descriptive statistics were calculated for demographic and other key variables. A series of four regression models were performed according to mediation analysis procedures (25, 26). Each regression model was tested using mixed model multivariate linear regression, accounting for the study's nested design and intra class correlation. All regression models were adjusted for gender, race/ethnicity, age, and socio-economic status. All analyses were conducted using SAS statistical software (version 9.1, 2003, SAS Institute, Carey, NC). The Sobel Test, a statistic that verifies the results of a mediation model, was used to evaluate for statistical significance (32, 33).
Among student participants, 52% were male; 63% were low income; and 61% were racial/ethnic minorities. For demographic and other key variables, there were no significant differences by gender. Participants reported a mean fruit/vegetable consumption of 3.6 servings per day (range: 0-24; Table 1).
Figure 1 depicts results of the mediation analyses. In the first regression model, self-efficacy to eat healthy was found to be significantly and positively associated with fruit/vegetable consumption (b1=0.13 p=0.026). In regression model two, the association between self-efficacy to eat healthy and perceived barriers to healthy eating was found to be significantly and negatively associated (b2=-0.15 p<0.001). The third regression model found that perceived barriers to healthy eating and fruit/vegetable consumption were significantly and negatively associated (b3=-0.45 p=0.002). Finally, when self-efficacy to eat healthy and perceived barriers to healthy eating were included in the same model, the association between self-efficacy and fruit/vegetable consumption was no longer significant (b4= 0.08, p=0.15), fulfilling the requirements for full mediation. The Sobel Test confirmed that perceived barriers fully mediated the association between self-efficacy and fruit/vegetable consumption (Sobel test statistic =2.7, p=0.007). Students in the study sample reporting higher self-efficacy had fewer perceived barriers and higher fruit/vegetable intake.
Other studies that have examined the association between self-efficacy and barriers to fruit/vegetable consumption have reported similar results (6, 34, 35). Previous studies have identified self-efficacy as a strong correlate of fruit/vegetable consumption across the lifespan (35, 36). In a recent review of the literature, Brug and colleagues found that positive self-efficacy to eat fruits/vegetables was significantly associated with daily fruit/vegetable intake among a sample of preadolescents (37). The current study investigated a more complex pathway via mediation as a means to explain the association between self-efficacy and fruit/vegetable intake.
Barriers to eating healthy as an important driver of adolescent fruit/vegetable consumption is supported by current literature. In a study of over 3800 adolescents, Lytle and colleagues reported that the greater the perceived barriers, the fewer fruits/vegetables consumed (38). Focus group research conducted among alternative high school students in Minnesota (39) found that perceived barriers, especially barriers related to access was an impediment to regular fruit/vegetable consumption. Adolescents felt that if fruits/vegetables were more readily available in their environment, such as in vending machines at school or on the table at home, they would be more likely to consume them. Other work conducted in alternative high schools highlighted that participation in school lunch programs was low due to “cold, unappealing meals.” At the same time, vending machines were well-stocked with salty, sugary snacks and beverages (40). Improving access to fruits/vegetables at school will be an important obesity prevention strategy. When environmental supports are in place, healthy choice becomes possible and self-efficacy has the opportunity to improve.
The majority of the study sample was low-income. Research also indicates that low-income students have greater barriers to fresh fruits/vegetables (41). Often, low-income families have to make the choice of spending their food dollar on inexpensive, calorie-dense foods rather than fruits/vegetables in order to make food stretch through the month (5, 22-24, 41). It has been demonstrated that if barriers such as price are removed, fruit/vegetable consumption increases (42, 43). For example, in a school-based intervention, it was found that pricing strategies to lower the cost of fruits and vegetables increased fruit/vegetable consumption for high school youth (43). The current findings suggest that the removal of price barriers may enhance students' self-efficacy to eat more healthfully and result in increase consumption of fruits/vegetables.
There are strengths and limitations to this study. To our knowledge, this is the first study to examine perceived barriers to healthy eating as a mediator of the association between self-efficacy to eat healthy and fruit/vegetable consumption, providing important information on the health behaviors of alternative high school students. Reliable measures were used to assess variables of interest. Given the nature of self-reported data, there was also potential for social desirability bias, such as overestimating fruit/vegetable intake (44, 45). Although the response rate is considered low, the study sample was representative of students attending the study schools, and similar to data describing students attending alternative high schools nationwide (11-14). Because only schools in the Minneapolis-St Paul area were included, the generalizability of study findings may be limited. Further, the small sample size may have limited the ability to detect significant differences in certain variables, such as race/ethnicity. As a cross-sectional study, the findings described are preliminary and cannot address causality. However, the mediation analysis did produce expected results, which supports further testing of the hypothesis in longitudinal and intervention studies.
There is an ongoing need for dietitians and other health professionals to promote increased consumption of fruits/vegetables among school-aged youth. Data indicate that students attending alternative high schools, a population of older youth, who are disproportionately low-income, minority and at an increased risk for overweight/obesity, may require additional support. The current study suggests that by addressing perceived barriers to healthy eating among students, self-efficacy to choose healthy foods may improve and fruit/vegetable consumption may increase. Further study is needed in alternative high schools in order to improve our understanding of the dietary behaviors of diverse youth and to help guide interventions that aim to meet the needs of this population.
The authors thank the school staff and students who participated in the Team COOL pilot study.
This research was supported by a grant from the National Institutes of Health/National Institute of Diabetes and Digestive and Kidney Diseases no. R21DK072948.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Meg Bruening, Nutrition PhD Student, Maternal Child Health Nutrition Trainee, Division of Epidemiology and Community Health, University of Minnesota, 1300 South Second Street, Suite 300, Minneapolis, MN 55454, Tel: 612.626.8984, Fax: 612.624.0315.
Martha Y. Kubik, Associate Professor School of Nursing, University of Minnesota, 6101 Weaver Densford Hall, 308 Harvard St SE, Minneapolis, MN 55455, Tel: 612.625.0606, Fax: 612.625.7091.
DenYelle Kenyon, Assistant Scientist, Health Disparities Research Center, Sanford Research/USD, 900 W. Delaware St., Sioux Falls, SD 57104, Tel: 605-328-0408, Fax: 605-328-0401.
Cynthia Davey, Senior Research Fellow, Biostatistical Design and Analysis Center, Clinical and Translational Science Institute, Academic Health Center, University of Minnesota, 717 Delaware St. SE, 223-21, Minneapolis, MN 55414, Tel: 612.624.7656, Fax: 612.625.2695.
Mary Story, Professor, Division of Epidemiology and Community Health, University of Minnesota, 1300 South Second Street, Suite 300, Minneapolis, MN 55454, Tel: 612.626.8801, Fax: 612.624.0315.