Nutritional PSA's on Fruit and Vegetable Consumption

This project will be carried out in three distinct phases: 1) development of commercials, 2) assessing food selection and consumption before, during and after commercial airing, and 3) outcome analysis and dissemination of results. Four elementary schools in Alachua County, Florida will be chosen to participate in this study. Two schools will have high participation in free and reduced fee school lunch programs while two will have lower participation. The schools will be divided such that two participate in the intervention (one with high free/reduced lunch participation and one with low) and two will serve as controls. All schools have a morning broadcast of announcements where 15-30 second fruit and vegetable "commercials" can be shown to all students in K-5th grades. Given the scope (1 year, limited budget) of this pilot project, we will assess purchasing and consumption behaviors of 3rd, 4th, and 5th graders in each school, rather than all children. As students at these schools (and the controls) have the option of what fruits and vegetables to select (if any) as part of the school lunch, the investigators will be able to observe if the commercials have an impact on student choice of healthy products. Students will be assessed collectively per class and thus no individual identifying information will be collected.

Two commercials will be prepared for the schools using messages developed for this age group that feature children as the main characters. This approach captures the potential influence of peers. Potential 3-5th grade "actors" and adult actors will be recruited from local drama/playhouse venues. The two commercials will be shown each week in alternating weeks for a total of four weeks. Example: Week 1, commercial A shown; Week 2, commercial B shown; Week 3, commercial 1 shown; Week 4 commercial B shown. Child actors will likely not attend the test schools, but they and their parents will be informed that the commercials will be shown in two Alachua County Public Schools. Children will include an equal mix of girls and boys and approximately half of the children will be from racial/ethnic minority groups that are represented in the Alachua County school system. These videos will be created using existing video production facilities at the University of Florida.

Six observations will be made at each school to assess the frequency of selection as well as the consumption of fruits and vegetables: once prior to the airing of the commercials, one day during each of the four consecutive weeks when commercials are shown, and once two weeks after the commercial airing has ended. This last assessment will allow the investigators to gather pilot information on whether any lasting effect of the commercials has occurred. At each school, all observations will occur on the same day of the week and children in grades 3-5 will be monitored. The frequency of fruit and vegetable purchase by children will be assessed using a combination of information from student identification numbers at the cash registers and University student monitors to observe the quantity of students selecting the fruit/vegetable promoted in the commercials. To assess food intake, the investigators will photograph each student's tray for future visual plate waste analysis (Silver 2008). Visual plate waste analysis allows trained student and faculty assessors to estimate whether 10, 25, 50, 75, or 90% of a food item remains uneaten on the plate. To evaluate the agreement between selection and consumption, a food item will be considered "consumed" if no more than 25% of the item remains on the plate. Each plate will be evaluated by a minimum of three trained assessors. If the values vary, the item will be discussed as a group until consensus is reached. Moreover, the investigators will work closely with the school food service directors in selecting our days for evaluation and ensuring that equal and adequate amounts of fruits and vegetables are offered on study days.

The primary outcomes are fruit and vegetable selection and consumption. With over 11,500 school students enrolled in the Alachua County School System at 23 elementary schools, we anticipate a total sample size of approximately 1000 3rd, 4th, and 5th grade children. Anticipating a 15% attrition rate over the study period, due primarily to absences, the investigators expect having repeated observations from 850 children, half from intervention schools, half from control schools. Although pilot in nature, this sample size is adequate to detect a mean difference fruit/vegetable selection frequency of 0.5 servings between the study schools with over 99% power using a two-sided 0.05 significance level to detect a standardized effect of 20 standard deviations. Whether food selection can serve as a proxy for food consumption as assessed by visual plate waste analysis will be assessed using regression analysis.