735 - Associations between objective fruit and vegetable consumption measured using the Veggie Meter and child weight in school-age children

Publication Number: 735.4019

Mary Jo Messito, New York University Grossman School of Medicine, New York, NY, United States; Christina N. Kim, New York University Grossman School of Medicine, New York, NY, United States; Sarah Lee, State University of New York Downstate Medical Center College of Medicine, Brooklyn, NY, United States; Carol Duh-Leong, Hassenfeld Children's Hospital at NYU Langone, New York, NY, United States; Michelle W. Katzow, Cohen Children's Medical Center, Northwell Health, New Hyde Park, NY, United States; Rachel S. Gross, New York University Grossman School of Medicine, New York, NY, United States

Background: Although adequate fruit and vegetable consumption (FVC) is associated with improved cardiovascular and metabolic health and lower risk of obesity at all ages, few children meet recommended FVC guidelines. While numerous interventions have been developed to improve FVC, it is time consuming and difficult to measure, and both parent and self-report methods can be inaccurate. Skin carotenoid levels measured using the Veggie Meter (VM) have been proposed as an objective measure of fruit and vegetable consumption because they correlate with reported FVC, but few studies have examined the associations between child weight and FVC using the VM.

Objective: 1) To assess FVC using the VM in school-age children, and 2) To explore associations between VM and child weight.

Design/Methods: Cross-sectional analysis using data from the 8-year old follow-up assessment of the Starting Early Program (StEP) randomized control trial of a primary-care based child obesity prevention intervention. StEP was developed for low-income, Hispanic mother-child pairs, beginning in the third trimester of pregnancy and continuing to child age 3 years. The VM score was obtained as an average of 3 skin -carotene measurements giving a continuous score of 0-800 where higher scores indicate higher FVC consumption. VM score of ≥280 represents adequate FVC. Child weights were obtained from direct measurement and medical record review, weight for age z-scores (WFAz) and ≥85th percentiles (overweight status) were calculated using CDC standards. Associations between VM score and child WFAz and overweight status at age 8 years were assessed using t-test, chi-square, multiple linear and logistic regression analyses, controlling for confounders (e.g., maternal pre-pregnancy weight status, education, non US-born, parity and intervention status).

Results: Among 242 children with weight and VM scores (Table 1), the mean VM score was 238 (SD=93.3), with 66 children having adequate FVC (≥280 VM). In bivariate analyses, children with adequate FVC had lower WFAz (0.67 vs 1.15, p=.01) and overweight status (38% vs 56%, p=.01); this difference remained significant after adjusting for confounders (Table 2). In adjusted and unadjusted regression analyses, higher VM score was associated with lower child WFAz (B=-.002, p=.05; B=-.002, p=.06) and lower rates of overweight status (B=-.003, p=.03; B=-.003, p=.04)

Conclusion(s): Higher FVC was associated with lower weight and rates of overweight status. The VM is a feasible objective way to measure FVC in school-age children, and can be used to assess FVC and impacts of interventions designed to improve it.

Family and Child Characteristics at Child Age 8 Years
table 1.pdf

Associations between Adequate Fruit and Vegetable Consumption by VM score, Weight-For-Age Z-Scores and Weight Status at Age 8 Years
table 2.v2.pdfAdequate FVC = VM score ≥ 280; Low FVC = VM score <280. The difference in WAFz was assessed using a t-test, and the difference in weight status was assessed using a chi-square test. Adjusted p-value obtained using ANCOVA and multiple logistic regression controlling for maternal education (did not complete high school vs. completed high school), country of origin (not born in the US vs. born in the US), parity (first child vs. not first child), and pre-pregnancy obesity (BMI <30 vs. BMI≥30).