(OR15-05-26) Association Between Veggie Meter Scores and Dietary Recalls in a Diverse Low-Income Urban Sample: Baseline Findings From the FRESH Study

Objectives: Accurate measurement of fruit and vegetable intake remains a challenge in population research. The Veggie Meter (VM) measures skin carotenoids using reflection spectroscopy and may provide a rapid, objective indicator of intake. This study examines associations between VM scores and two-day self-reported intake of total fruits and vegetables, total vegetables, and red/orange vegetables measured by ASA24.

Methods: This cross-sectional analysis used baseline data from customer participants (N=236) in the FRESH restaurant intervention study conducted in Baltimore, MD (n=76) and the Washington, DC metropolitan area (n=160). Restaurants were recruited from low-income census tracts ( >15% below poverty) meeting racial/ethnic criteria ( >80% non-Hispanic Black in Baltimore; >35% Latino in DC). Customers (≥1 visit/week) were recruited via intercept sampling. Participants completed Veggie Meter measurements, two ASA24 dietary recalls administered by trained data collectors, and demographic/anthropometric questionnaires. Associations between VM scores and dietary intake were assessed using Spearman correlations and multivariable linear regression adjusting for age, sex, race/ethnicity, and BMI.

Results: Among 236 participants, 55.5% were female, with a mean age of 43.9 ± 15.0 years and BMI of 30.7 ± 7.5 kg/m². The DC sample was largely Hispanic/Latino (93.8%), while the Baltimore sample was primarily non-Hispanic Black (84.2%). Mean intake was 3.29 cups/day for total fruits and vegetables, 1.98 cups/day for total vegetables, and 0.45 cups/day for red/orange vegetables. Mean Veggie Meter score was 255.5 ± 91.1 and was higher in DC than Baltimore (273.8 vs 217.0); higher scores reflect greater skin carotenoid status. VM scores were positively correlated with total vegetables (ρ=0.213, p=0.001), red/orange vegetables (ρ=0.205, p=0.002), and total fruits and vegetables (ρ=0.134, p=0.039). In adjusted models, only total vegetable intake remained associated with VM scores (β = 10.10 per additional cup/day, 95% CI: 0.60-19.61).

Conclusions: VM scores were modestly associated with reported vegetable intake in this low-income urban sample. Reflection spectroscopy may complement dietary recalls as a rapid, low-burden tool for assessing vegetable intake.

Funding Sources: NIH, National Institute on Minority Health and Health Disparities