Source: Fairleigh Dickinson University submitted to
PREVENTION OF LATE ADOLESCENT OBESITY IN THE COLLEGE ENVIRONMENT: AN OPTIMAL DEFAULT PARADIGM
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
NEW
Funding Source
Reporting Frequency
Annual
Accession No.
0231957
Grant No.
2013-69001-20416
Project No.
NJW-2012-03386
Proposal No.
2012-03386
Multistate No.
(N/A)
Program Code
A2101
Project Start Date
Jan 1, 2013
Project End Date
Dec 31, 2014
Grant Year
2013
Project Director
Loeb, K. L.
Recipient Organization
Fairleigh Dickinson University
1000 River Road T-BE2-02
Teaneck,NJ 07666
Performing Department
School of Psychology
Non Technical Summary
The first year of college is a time of transition when adolescents are first learning to manage their own energy intake and expenditure without parental supervision. Many students obtain their meals on campus and are confronted with a wide array of food choices without any limitation on options or portion sizes. Not surprisingly, researchers have documented a consistent weight gain in first-year undergraduates averaging 1-4 kg, resulting in a substantially greater proportion of freshman classified as overweight at the end of the first semester (25.2%) than the beginning (15.6%).Thus, the first year of college is considered a critical period for weight gain for adolescents and consequently a target for prevention. Systemic changes in the food delivery systems on college campuses have the potential to reach a large number of students with an overall low cost, if they can be shown to be effective in preventing obesity. Obesity prevention symbolizes a core philosophical tension between individual freedom and government/institutional actions designed to promote public welfare. Manipulations of available options, such as default choices in decision contexts, represent a compromise between the two; public policies can determine the optimal default positions, yet the choice remains with the individual to opt out. The proposed study will be the first to test the concept of optimal defaults around obesogenic behaviors in college students. By systematically manipulating the health value of the default mode, the experiment in this study examines how hard individuals will work to remain in or obtain healthy choices in the context of the college dining system. Specifically, we will randomize participants to one of three conditions in an experimental meal paradigm: (a) an optimal default lunch, (b) a sub-optimal default lunch, or (c) a free-choice array lunch. We hypothesize that making the default options more optimal (less obesogenic) will lead to more frequent choice of healthier foods (yielding less caloric intake, and more fruit and vegetable consumption) than either the suboptimal default or free array lunches. A major long-term goal of the body of proposed current and future work is sustainable institution-based policy implementation regarding food that will extend beyond the duration of the project. Specifically, results of these experiments will inform policies regarding default selection procedures in college and university dining programs.
Animal Health Component
100%
Research Effort Categories
Basic
(N/A)
Applied
100%
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
70360103070100%
Knowledge Area
703 - Nutrition Education and Behavior;

Subject Of Investigation
6010 - Individuals;

Field Of Science
3070 - Psychology;
Goals / Objectives
This project examines how the manipulation of the default mode (i.e., the usual and customary decisions that are made preemptively for the public unless they opt out) may serve as a critical strategy in the prevention of adolescent obesity. By systematically manipulating the health value of the default mode, the experiment outlined in this proposal examines how hard individuals will work to remain in or obtain healthy choices in the context of the college dining system. Specifically, this study aims to generate new knowledge about the behavioral and environmental factors that influence excessive weight gain among youth by studying the effects of optimal (healthy) defaults in the dining experience of first year college students, a group at high risk for weight gain. We will randomize participants to one of three conditions in an experimental meal paradigm: (a) an optimal default lunch, (b) a sub-optimal default lunch, or (c) a free-choice array lunch. Age- and gender- adjusted calories consumed will be the primary outcome variable. The secondary outcome variable will be quantity of fruit and vegetable consumption (servings). We hypothesize that making the default options more optimal (less obesogenic) will lead to more frequent choice of healthier foods (yielding less caloric intake, and more fruit and vegetable consumption) than either the suboptimal default or free array lunches. The first 6 months of the grant will be spent (a) designing the specific menus for the experiments and calculating their precise caloric contents; (b) conducting palatability checks of the menu items with first year college students; (c) developing the demographic questionnaire and food preference inventory specific to this study; (d) procuring and designing materials for the experiment; (e) interviewing and hiring graduate student research assistants; (f) training research assistants and establishing inter-rater reliability in food measurement methods; and (g) developing recruitment materials and actively recruiting for the study. The following 12 months will be spent running the weekly experiments and entering data. The final 6 months of the study will be spent completing data entry, conducting data analysis, writing up the experiments for publication, and disseminating the findings via conference presentations and to college administrators and communities. The fundamental research from this pilot study will inform several aspects of a future Integrated Research, Education, and Extension to Prevent Childhood Obesity proposal, including: (a) expansion of college-based optimal default experiments to include energy expenditure (exercise) in addition to energy intake paradigms, (b) applied research in the form of a college-wide preventive intervention for late adolescent obesity that incorporates optimal defaults in the dining environment, and uses a longitudinal design, (c) teaching and training for doctoral-level graduate students in obesity prevention research and strategies, and (d) targeted outreach efforts to other colleges to shift default health-related environmental conditions to be more optimal.
Project Methods
We will conduct a randomized, between-subjects, 3-condition experiment to test the dietary effects of optimal vs. suboptimal vs. free choice array food presentations in first year college students' dining experience. The sample will consist of 135 first year college students. Data will be analyzed with a one-way ANCOVA to test for a main effect of condition on calories consumed, with participant gender and BMI entered as covariates. In addition, we will use the Tukey Post-Hoc test to examine pair-wise comparisons between any two of the three conditions if the overall model is significant. This same approach will be used to test our secondary outcome variable, servings of fruits/vegetables consumed. Participants will receive a standardized pre-packaged pre-load breakfast (to minimize variability in hunger/satiety across participants) following the informed consent process, measurement of BMI on a physician's balance scale with stadiometer, and completion of a demographic questionnaire and measures to assess predictor variables. They will return at lunchtime for the experimental meal in a private dining room. In the optimal default condition, the foods readily available will be healthy, displayed in pre-packaged individual servings. Healthy will be defined as high nutrient, low energy-dense items. There will also be a menu in a stand on the table that lists the foods available in large print; at the bottom of the menu, in smaller print, a list of alternate, unhealthy (lower nutrient, high energy-dense) items will be noted to be available, but requiring a 15 minute wait to obtain. Participants may choose any combination of items from the default display or the alternate menu, to mimic the allowance of choice typical in a college dining hall. All food items will be pre-measured and calculated for macronutrient analysis. Selected foods will be noted in a check-out station. Participants will be instructed to leave their trays at the conclusion of their meal, and the researchers will calculate percent of food left on the plate, and calories and fruits and vegetables consumed by subtracting leftovers from the checked-out items. We will train research assistants in evidence-based methods to rate percent of food leftover and establish good inter-rater reliability prior to the study. We will use an energy density cut-off of 1 kcal/gram for fruits and vegetables as our secondary dependent variable. In the sub-optimal default condition, the conditions will be reversed in that the default lunch will consist of the less healthy items, and participants will have to actively opt out of such items and wait for the healthier options if desired. In the free array condition, both healthy and unhealthy items will be displayed simultaneously, with no wait for any item. This represents the control condition, most similar to the standard college dining environment and mechanisms of choice within. Final steps of the project include disseminating the findings via conference presentations and to college administrators and communities to inform shifts in default health-related environmental conditions to be more optimal.

Progress 01/01/13 to 12/31/13

Outputs
Target Audience: Our ultimate target audience is the general population of late adolescents in the high-risk period of the first year of college. In particular, we are interested in modifying the college food environments of segments of this population that are most vulnerable to childhood obesity, to address the significant racial, ethnic, and socioeconomic disparities in obesity prevalence among US adolescents. The ethnic identity of the participants in the study is 47% Hispanic/Latino; the racial identity is primarily black or African American (23%), followed by white (17%), multiracial (13%), Asian (7%), and American Indian or Alaskan Native (3%), with 37% not reporting a racial identity. This breakdown reflects the diversity typical of students at our university. When the experiment is complete, we will distribute debriefing information to participants about the optimal default paradigm as it applies to adolescent obesity prevention and the specific hypotheses of this study. Ultimate extension and outreach efforts extending from this research-focused seed grant project would target colleges and universities serving a diverse student body at elevated risk. Changes/Problems: We are approximately two-thirds of one semester behind on our research schedule. The original study timeline called for the experiment commencing at the start of the fall 2013 semester. However, a fire on campus shut down one building for several months, and some classes were temporarily re-situated in the space originally reserved for our study, until more than halfway through the fall semester. In addition, there were several weather-related closures this winter that required re-scheduling experimental sessions. In the original grant proposal, under a section titled, “Expected Outcomes, Limitations, and Potential Problems,” we planned for the following contingency: “If recruitment challenges arise, time can be made up by running more participants (>5) on a given day or extending recruitment into a third semester within the timeframe of the grant.” While our challenge in this case was not recruitment-related, we are implementing both of these options (running concurrent experimental sessions and recruiting through the fall 2014 semester) to compensate for lost time during fire- and weather-related emergency building and university closures. What opportunities for training and professional development has the project provided? While training and professional development were not specific goals of this seed grant, we have provided training and mentorship on childhood obesity prevention in the context of work on this project to six PhD candidates in Clinical Psychology and one undergraduate Psychology major at our university. How have the results been disseminated to communities of interest? As the experiment is still in progress, there has not yet been dissemination of results. As noted above in Accomplishments, activities planned for the final stages of the grant period and just after include writing up the experiments for publication, and disseminating the findings via conference presentations and to college administrators and communities. A major long-term goal of this work is sustainable institution-based policy implementation regarding food that will extend beyond the duration of the project. As described above in Other Products, we have presented the theoretical model of optimal defaults as applied to the prevention of childhood obesity, and the proof-of-concept experiment incorporated in this seed grant (not results), in invited talks. Slides from the talk at the Yale University Rudd Center for Food Policy and Obesity in September 2013 are accessible to the public at http://www.yaleruddcenter.org/news.aspx?id=241 and a podcast interview after the talk is available to the public at https://soundcloud.com/#yaleuniversity/two-strategies-to-change. What do you plan to do during the next reporting period to accomplish the goals? The majority of the remaining time on the grant will spent running the weekly experiments and entering data. As described below in Changes/Problems, we are running concurrent experimental sessions and recruiting through the fall 2014 semester to compensate for lost time during fire- and weather-related emergency building and university closures. As noted above in Accomplishments, the final stages of the grant period and just after will be spent completing data entry, conducting data analysis, writing up the experiments for publication, and disseminating the findings via conference presentations and to college administrators and communities.

Impacts
What was accomplished under these goals? The first year of college is a time of transition when adolescents are first learning to manage their own energy intake and expenditure without parental supervision.Many students obtain their meals on campus and are confronted with a wide array of food choices without any limitation on options or portion sizes.Not surprisingly, researchers have documented a consistent weight gain in first-year undergraduates averaging 1-4 kg, resulting in a substantially greater proportion of freshman classified as overweight at the end of the first semester than the beginning. Thus, the first year of college is considered a critical period for weight gain for adolescents and consequently a target for prevention. Systemic changes in the food delivery systems on college campuses have the potential to reach a large number of students with an overall low cost, if they can be shown to be effective in preventing obesity. This project examines how the manipulation of the default mode (i.e., the usual and customary decisions that are made preemptively for the public unless they opt out) may serve as a critical strategy in the prevention of adolescent obesity. By systematically manipulating the health value of the default mode, the experiment in this grant examines how hard individuals will work to remain in or obtain healthy choices in the context of the college dining system. Currently, most societal defaults are set to a suboptimal or deficient capacity with regard to obesity prevention. For example, unhealthy foods are more available, accessible, and affordable than healthy foods in poor neighborhoods, making it easier to choose the former over the latter in day-to-day food choices. In college dining environments, catering to student preferences means that the less healthy foods are easily visible and readily obtained. In an optimal default model, healthier foods would be the default, and students would have to exert effort and tolerate some inconvenience to access alternatives. Research in multiple domains suggests that even when free choice is preserved, individuals are highly likely to remain with the choice presented to them as the default. This study is the first to test the concept of optimal defaults around obesogenic behaviors in college students. We are randomizing participants to one of three conditions in an experimental meal paradigm: (a) an optimal default lunch, (b) a sub-optimal default lunch, or (c) a free-choice array lunch. We hypothesize that making the default options more optimal (less obesogenic) will lead to more frequent choice of healthier foods (yielding less caloric intake, and more fruit and vegetable consumption) than either the suboptimal default or free array lunches. The project director (PD), co-PD, and their research team, including collaborating consultants, have set up the experiment and run 30 participants to date, with 30 more participants identified and pending (the proposed ultimate sample size is 135). Specifically, the first period of the grant was spent on the following activities: (a) designing the specific menus for the experiments and calculating their precise caloric contents; (b) developing the study-specific demographic questionnaire and food preference inventory (which was used for palatability checks with our first set of participants); (c) procuring and designing materials for the experiment; (d) interviewing and hiring graduate student research assistants; (e) training research assistants in all study procedures, including reliably measuring and serving specific portions of each menu item and obtaining pre- and post-meal food weights using a research-quality food scale; (f) developing recruitment materials and actively recruiting for the study, and (g) running the experiment. Preliminary results on the first set of participants (n=30) show that when the default meal is optimal, 100% of the food selected represents healthier meal choices (e.g., baked chicken breast, brown rice pilaf, green salad with fat-free dressing, fresh fruit) even though alternative items are listed on the menu and are available by request and with a brief (15-minute) wait. Conversely, when the default meal is sub-optimal, 100% of the food selected represents obesogenic meal choices (e.g., fried chicken, fried rice, Caesar salad, pie), even with the clear option to access alternatives. When there is a free-choice array and all items are presented simultaneously, only two-thirds of the selected items were optimal, i.e., the healthier choices. The final stages of the grant period and just after will be spent completing data entry, conducting data analysis, writing up the experiments for publication, and disseminating the findings via conference presentations and to college administrators and communities (change in knowledge). A major long-term goal of this work is sustainable institution-based policy implementation regarding food that will extend beyond the duration of the project. If our hypotheses are supported with the final sample and proposed outcome variables, results of this experiment can inform policies regarding default selection procedures in college and university dining programs (change in action). Such changes in dining selection procedures can alter consumption, which in turn can prevent increases in weight status from normal to overweight, or overweight to obese during this high-risk period of development (change in condition).

Publications

  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2014 Citation: Loeb, K. L., Radnitz, C., Keller, K., Schwartz, M., Boutelle, K., & Marcus, S (2014, July). Prevention of late adolescent obesity in the college environment: An optimal default paradigm. Poster to be presented at the meeting of the Society for Nutrition Education and Behavior, Milwaukee, WI.