The purpose of this study was to examine the effects of a cooking and gardening nutrition education intervention on health behavior mediators and diet quality, while also seeking to examine the interaction between health behavior mediators and their effect on self-efficacy and fruit and vegetable intake in food insecure college students. The intervention was successful in improving certain health behavior mediators, such as self-efficacy, reciprocal determinism, and personal health views, as well as fruit and vegetable intake. The analysis also demonstrated that reciprocal determinism had a direct effect on improving self-efficacy of healthy behaviors in this population.
HEALTH BEHAVIOR MEDIATORS
Nutrition knowledge and outcome expectation. At the conclusion of the study, those who participated in the study significantly increased personal health views, reciprocal determinism, and self-efficacy. However, nutrition knowledge and outcome expectations did not significantly increase in the intervention group when compared to the control. When investigating why nutrition knowledge scores did not change, the content of the nutrition education program was compared to the specific questions in the nutrition knowledge questionnaire. While the questionnaire was validated and relevant to this demographic, it was found that the questionnaire content was not congruent with the materials being taught within the sessions. The sessions taught general information on MyPlate and food groups while the questions on the questionnaire were mostly related to specific food items and their effect on health. Upon analysis, the average correct response for the one relevant question, ‘a healthy meal consists of half meat, a quarter vegetable, and a quarter side” did significantly increase in the intervention group compared to the control group. Currently, no studies have been published on the effects of nutrition interventions on nutrition knowledge in food insecure college students. Previous education interventions on college athletes and the general college population have successfully improved nutrition knowledge, which can lead to improved dietary behaviors.35, 36
There was also no change found for outcome expectation scores, which were unusually high at baseline.7,11 This left little room for improvement post intervention. Previous nutrition education studies using the SCT have improved these scores and increased the understanding of the long-term effects of health in young adults.7,11
Personal health views. This study successfully improved personal health views in food insecure students who completed the intervention. By improving personal health views, these results indicate that the behaviors practiced through the intervention helped students feel healthier. Few studies have directed interventions at improving self-perceived health even though personal health views have been shown to have a direct effect on dietary behavior and on intention to create behavior change in adults and low-income populations.37,38 In college students, negative health views have been associated with increased weight gain.39 Improving personal health views may encourage food insecure students to continue practicing healthy behaviors which may improve diet and overall long-term health.
Reciprocal determinism. Reciprocal determinism was also improved in those who participated in the intervention. In this study, improving reciprocal determinism indicates that the intervention was successful at helping students create healthier environments for themselves. To date, previous nutrition intervention studies have not directly assessed changes in reciprocal determinism. In this study, gardening was used to improve the participants direct environment, improve reciprocal determinism, and ultimately improve healthy behaviors. It is hypothesized that the act of gardening increases fruit and vegetable intake through an exposure effect. By improving their environment and increasing reciprocal determinism, the study essentially also increases exposure to fruits and vegetables, supporting this concept. This indicates that those who surround themselves with, and actively participate in the care and procurement of, fruits and vegetables are more likely to create positive associations and consume more fruits and vegetables. Overall, by improving both a student’s direct environment and their personal beliefs, they are more likely to adopt healthier lifestyle behaviors which may improve their short-term and long-term health.
Self-efficacy. Lastly, an increase in self-efficacy indicates that the intervention helped students feel more confident in their ability to adopt healthier diets. Alternately, studies have directly assessed change in self-efficacy in college students, as it is a key construct in the SCT for changing health behaviors.9,22,40,41,42 These studies, which have all incorporated either gardening or cooking with nutrition education, have successfully improved self-efficacy of health behaviors. 9,22,40,41,42 When determining the effect of cooking interventions in college students, previous studies have successfully increased self-efficacy, knowledge, and behaviors associated with cooking and healthy behaviors.21,26,44-46 Specifically, studies that have utilized interactive experiences, like those utilized in the current study, have been shown to be more effective at improving self-efficacy than those using other modalities.21,43,45
Effect of mediators on self-efficacy. Since self-efficacy has proven to have such a large impact on behavior directly, the study aimed to investigate the relationship between the SCT mediators used in this intervention on self-efficacy. Only changes in reciprocal determinism were significantly correlated with overall change in self-efficacy. This is to say, by creating a healthy environment, food insecure college students believed they were able to follow a healthy diet. Specifically, this study used gardening and the accessibility to fresh herbs and vegetables to create a healthier environment. As mentioned earlier, while no studies have specifically looked at the effects of reciprocal determinism or changes in environments on self-efficacy, various studies using the SCT as a whole have seen positive changes in self-efficacy, including those using gardening and cooking interventions.43,45-49 While nutrition knowledge and outcome expectations did not have a significant relationship with self-efficacy, neither significantly changed in the intervention group. It is possible that a more accurate measure of nutrition knowledge change in the current study might correlate differently with the change in self-efficacy. Overall, these results indicate that by teaching gardening, cooking, and nutrition skills to food insecure college students, they will view themselves as healthier, create healthier environments for themselves, and increase their self-efficacy for healthy behaviors.
FRUIT AND VEGETABLE INTAKE
Overall, the intervention successfully improved fruit and vegetable intake in food insecure college students. Compared to those who were in the control group, students who were part of the intervention did increase their fruit and vegetable intake. The results of this study are supported by previous literature which has shown that, although separately, nutrition education and gardening can successfully increase fruit and vegetable intake in college students.7,20,50 Currently, only one previous nutrition education study has successfully increased fruit and vegetable intake in the college population.23 Many other nutrition education intervention studies have been conducted in college students but these have only been successful in improving other dietary behaviors, such as increasing low fat dairy intake and reducing consumption of sugar sweetened beverages.9,43,51,52
Gardening has also previously been associated with increasing fruit and vegetable intake in multiple populations.53,54 In college students specifically, Loso et al. and Staub et al. have directly associated gardening with increased fruit and vegetable intake. In comparison to the current study, these studies did not actively apply a gardening intervention but assessed self-reported gardening reports in comparison to fruit and vegetable intake.20,50
As stated previously, cooking interventions in college students have been successful in improving self-efficacy for healthy behaviors but have not previously been successful at increasing fruit and vegetable intake. Since food insecure students have additional needs, combining cooking with nutrition education and gardening can potentially be more impactful to this at-risk group compared to the than to the general college population. The results from this study have shown that a hands-on, multi-faceted intervention can successfully improve fruit and vegetable intake in food insecure college students.
Effect of mediators on fruit and vegetable intake. Finally, the study aimed to determine whether SCT health mediators would be correlated with change in fruit in vegetable intake in those who participated in the intervention. It was determined that theory mediators had no correlation with the change in fruit and vegetable intake. The intervention did not improve nutrition knowledge and outcome expectation scores and thus these would not have an effect on changing fruit and vegetable intake. Self-efficacy and reciprocal determinism which ultimately did improve post intervention, were also not associated with change in fruit and vegetable intake in this study. Previous nutrition education studies have linked self-efficacy, specifically, to changes in fruit and vegetable intake in nutrition education interventions in college students.7,9
The lack of significance and correlation between the mediators and fruit and vegetable intake in this study is likely due to the small sample size. While it was hypothesized that specific mediators would have an effect on change in fruit and vegetable intake, it is plausible and likely that the change in fruit and vegetable intake arose from the mediators working together as whole or from other factors not considered, such as the combination of using three educational modalities (cooking, gardening, and nutrition education). The results from this study show that a health intervention specifically tailored for food insecure college students is effective at increasing in fruit and vegetable intake.
LIMITATIONS
This study has several strengths and limitations. Strengths included the use of a randomized control trial and the stratification of participants though BMI. This design ensured proper comparison among even distributed groups. The study also had a strong theoretical design using SCT to create the 6-week nutrition education program. When creating an intervention based on theory, the nutrition educator relies on a set of components that describe, predict, and explain behavior change which when addressed should result in a direct effect on what drives behavior change.55 Additionally, the SCT, specifically, has been shown in multiple studies to be successful in creating health related behavior changes.7,9,41,45 Another important strength of this study was the large population of Hispanic participants. Participants in previous studies focused on food insecure college students are largely White Caucasian.1 Previous literature has demonstrated that Hispanic and other minorities are at a greater risk for food insecurity.56-58 College food insecurity studies involving a larger variety of postsecondary institutions, including Hispanic-serving institutions, are important to better understand the severity and rate of college food insecurity.1 The results of this study demonstrate the ability to improve healthy behaviors in an at-risk population.
Limitations of this study include an uneven distribution of gender, a small sample size and the use of self-reported data. The study had a large majority of female participants which does not represent the general college population or the food insecure college population. Current literature estimates the gender distribution of food insecure college students matches that of the general college population, composed of 50-75% female.1 Additionally, the intervention group of the study consisted of only 42 students. While this sample was large enough to produce significant effects for outcome variables, a larger sample size would allow for more profound and generalizable conclusions. Additionally, the use of self-reported data can lead to inconsistencies and bias. Self-reported data in this study included the use of questionnaires to determine food insecurity, nutrition knowledge, and fruit and vegetable intake. It is important to note that all previous studies mentioned have assessed fruit and vegetable intake in a self-reported measure which allows results to be directly comparable to this study.
Future research should involve a larger sample and an even gender distribution to provide generalizable data on the effects of education interventions on health behaviors in food insecure college students. Large scale studies, including students from different college campuses should be included. Future intervention studies on food insecure college students should include a follow up or longitudinal design to determine whether an intervention can increase healthy behaviors long term. In addition, studies should also include students who are not food insecure, in a randomized control trial, to compare differences in the two groups and to determine if mediators, such as stress and life satisfaction, differ among the two groups as this might influence change in healthy behaviors. Finally, implementation of this program and additional resources across college campuses may improve college food insecurity rates and should be evaluated in the future.