Outcome Measures and Analysis: A 5-point Likert- scale-based questionnaire was administered to 37 students, and distribution of responses (%) was examined. Focus groups’ transcripts were subjected to content analysis. Results: Fifty-seven percent of teenagers agreed/strongly agreed that what their family thinks, says, or does incites them to make particular food choices, whereas the result was 27% for friends’ influence. According to participants, family norm was mainly positive. Other interpersonal in- fluences included trusted teachers and trainers. Most par- ticipants (65%) agreed/strongly agreed that family support was important to adopt healthy eating habits (friends’ support: 43%). The importance of the self-influ- ence emerged throughout the interviews. Conclusions and Implications: Family would have more influence on eating behaviors than friends. The de- velopment of supportive interventions targeting low-in- come families would be an asset in promoting healthy eating among disadvantaged teenagers. P90 Factors Influencing Japanese Children With High Scores on the ‘‘Gratitude for Food’’ Scale Rie Akamatsu, DrPH, akamatsu.rie@ocha.ac.jp, Ochanomizu University, 2-1-1, Otsuka, Bunkyo-ku, Tokyo, 112-8610, Japan Objective: Japanese people tend to believe that people should be thankful for their food because food is necessary for life, and wasting food is undesirable. This study de- scribes the characteristics of children with high scores on the ‘‘Gratitude for Food’’ scale, which measures thankful- ness for food. Design, Setting and Participants: The cross-sectional study was conducted with 2,070 fifth- and sixth-grade stu- dents at elementary schools in Tokyo, Japan. Outcome Measures and Analysis: Cross-tables were used to compare factors (eg, demographic characteristics, preferences and behaviors related to fruit and vegetables [F&V], gardening experience, mealtime routines at home) influencing the high- and low-scoring groups on the ‘‘Grat- itude for Food’’ scale. Logistic regression analysis was con- ducted on each factor to evaluate its association with the high-scoring group. Results: Children in the high-scoring group liked and ate more F&V. In addition, most said that they had had gar- dening experience; most of them also stated that their fam- ily advocated gratitude for food during mealtimes at home. With the exception of F&V behaviors, these variables were also associated with the high-scoring ‘‘Gratitude for Food’’ group in the logistic regression analysis. Conclusions and Implications: This study suggests that gardening experience and favorable mealtime situa- tions at home result in higher ‘‘Gratitude for Food’’ scores among children. The results also suggest ‘‘Gratitude for Food’’ possibilities related to encouraging children to eat more F&V. P91 Cancelled P92 Student Involvement in LWPs From Development to Implementation: Pennsylvania Case Study Lamis Jomaa, MS, luj111@psu.edu; Elaine McDonnell, MS, RD, etm101@psu.edu; Charles Orlofsky, BA; Claudia Probart, PhD, RD, Department of Nutritional Sciences, Pennsylvania State University, 110 C Chandlee Laboratory, University Park, PA 16802 Objective: The purpose of this study is to assess imple- mentation of student involvement goals in local wellness policies (LWPs) developed by Pennsylvania school districts (SDs). Student involvement in school nutrition policies is a promising area in obesity prevention. Design, Setting and Participants: LWPs submitted by SDs to the Pennsylvania Department of Education in 2006 were abstracted by our research group. In 2008, a policy implementation checklist (PIC) was developed to evaluate the implementation of LWPs. A total of 368 SDs that devel- oped LWPs and responded to PIC were included in the analysis. Outcome Measures and Analysis: Student involve- ment goals included in LWPs and reported on the PIC, and SES variable (percentage of students eligible for free and reduced-price meals %F/R]) were analyzed. Descrip- tive statistics and analyses of variance were run using SPSS. Results: Some measures of student involvement in- cluded in LWPs were not subsequently reported to be im- plemented, such as involving students in wellness committees or in annual reviews, assessing student in- puts, and providing them with nutritional content of school meals. SDs providing students with nutritional content of school meals and assessing student inputs had lower %F/R (F ¼ 6.6, P ¼ .011; F ¼3.3, P ¼ .07, respec- tively) than those SDs not involving students in these ways. Conclusions and Implications: Many SDs are not im- plementing student involvement goals previously in- cluded in their LWPs. Demographic and policy characteristics of schools may influence the level of stu- dent involvement and need to be explored. Child Nutri- tion Reauthorization 2010 is a unique opportunity to review the LWP mandate and emphasize student involve- ment in LWP implementation. This project is funded by the Robert Wood Johnson Foundation, Healthy Eating Re- search Program. P93 Effect of a Maternal Nutrition Education Intervention in the First Year of Infant Life on Dietary Quality in Childhood: A Randomized Controlled Trial Marcia Vitolo, PhD, vitolo@ufcspa.edu.br; Fernanda Rauber, nandarauber@hotmail.com, University of Health Sciences of Porto Alegre, 245 Sarmento Leite Street, Porto Alegre, P89 (continued) S122 Poster Abstracts Journal of Nutrition Education and Behavior  Volume 42, Number 4S, 2010 Continued on page S123