efﬁcacy, a positive relationship to food, calorie restriction, and body satisfaction. Results: All ﬁve dimensions of the MFL were signiﬁ- cantly and positively associated with fruit and/or vege- table intake, with moral and health showing the greatest associations; moral (fruit r ¼ .29 p < .001; vege- table r ¼ .35, p < .001) and health (fruit r ¼ .35, p < .001; vegetable r ¼ .40, p < .001). The health, sacred and moral meaning dimensions were signiﬁcantly and negatively associated with processed snacks and junk food con- sumption (both p < .05), and the moral dimension was signiﬁcantly and negatively associated with red meat consumption (p < .01). The moral, aesthetic and health domain were signiﬁcantly and positively associated with greater self-efﬁcacy for consuming healthy foods (all p < .001). Both the moral and health domains were signiﬁcantly and positively associated with body satisfac- tion (both p < .01). All ﬁve domains of the MFL were signiﬁcantly and positively associated with a positive relationship to food, and none were associated with cal- orie restriction. Conclusions and Implications: These data suggest that the MFL construct has clinical health relevance in the form of promoting healthier dietary behavior and a positive rela- tionship to food. Funding: None O14 Assessment of Impact on Objective Biometric Outcomes of the Eating Smart  Being Active Curriculum Benjamin Gowan, MS, benjamin.gowan@colostate.edu, Colorado State University, 102 Gifford 1571 Campus Delivery, Fort Collins, CO 80523; Garry Auld, PhD, Colorado State University; Susan Baker, EdD; Christopher Melby, DPH; Ann Hess, PhD Objective: To measure the impact of the Eating Smart  Being Active (ESBA) curriculum on objective biometric health outcomes. Design, Settings, Participants: A longitudinal data analysis using pre-post intervention and six-month follow-up data from adults (n¼61) completing Colorado EFNEP classes using the Eating Smart  Being Active curric- ulum during 2016-2017. Outcome Measures and Analysis: Biometric data collected included height and weight for Body Mass Index (BMI), Systolic and Diastolic Blood Pressure (SBP/DBP), and Hemoglobin A1c (HbA1C). One tail paired t-tests were used pre-post and repeated measures ANOVA for pre, post, follow-up analyses. Signiﬁcance was set at p < 0.05. Results: All biometric measures experienced an unad- justed mean decline from pre to post, however only DBP and HbA1C were signiﬁcant. Mean weight declined 0.95kg (p ¼ 0.17), mean BMI declined 0.24 kg/m2 (p ¼ 0.25), mean SBP declined 0.90 mmHg (p ¼0.10), mean DBP declined 1.72 mmHg (p¼0.01) and mean HbA1c declined 0.11 percent (p ¼0.04). Six-month follow-up measures are in progress and will be included in the anal- ysis in May 2017. Conclusions and Implications: This pilot study dem- onstrates reductions in objective biometric outcomes associated with participation in an EFNEP intervention using the widely adopted Eating Smart  Being Active cur- riculum, notably for blood pressure and blood glucose levels. The positive trends in biometric measures support programmatic conclusions based on participants’ self-re- ported behavior changes and provide initial objective ev- idence that EFNEP might positively affect health status. Further research is needed to improve evidence for causality. Funding: NIFA O15 Development and Validation of a Questionnaire to Assess Fruit and Vegetable (FV) Consumption at School Lunch Matthew M. Graziose, MS, mmg2198@tc.columbia.edu, Columbia University, 525 West 125th Street, Box 137, New York, NY 10027; Pamela A. Koch, EdD, RD, Columbia University; Randi Wolf, PhD; Heewon Lee Gray, PhD, RD; Isobel Contento, PhD, RD Objective: Develop and validate a modiﬁable, paper-and- pencil questionnaire assessing FV consumption at school lunch. Design, Settings, Participants: Five phase validation study, in which the questionnaire was iteratively modiﬁed. The setting was 23 public elementary schools from eight states in the U.S. (70% urban; 77% free/reduced price lunch eligible) participating in the National School Lunch Program. Second and 3rd graders (n¼41, 62, 71, 63, 976 for Phases 1-5, respectively) participated. Outcome Measures and Analysis: The Fruit and Vegetable Recall Questionnaire (FVRQ), adapted from Pax- ton et al (2011), was administered within the class- room directly after lunch and assessed FV items taken and amounts consumed. FVs were tailored based on lunch that day. A digital photography method was the referent; students’ trays were photographed in the cafeteria before and after eating and photos were visually analyzed. Agree- ment between methods was calculated as rates of matches, intrusions, and omissions for each Phase. Results: Questionnaire modiﬁcations included chang- ing page orientation, increasing font size, reducing response options, using smiley faces for scales of mea- surement, and adding a training page. The match rate for FV items taken was high (Phases 1-5: 82%, 82%, 91%, 92%, 89%, respectively), with intrusions more frequent than omissions. For amounts eaten, match rates were moderate, but improved throughout Phases 1-5 (68%, 67%, 84%, 83%, 76%); students more frequently overestimated than underestimated consumption. Sec- ond and third grade students had similar match rates for items taken (87% vs. 87%) and amounts consumed (72% vs. 75%). O13 (continued) S6 Oral Abstracts Journal of Nutrition Education and Behavior  Volume 49, Number 7S1, 2017 Continued on page S7