Abstract Research on the relationship between cancer risk and fruit and vegetable consumption is primarily focused on the quantity of consumption. However, simply looking at the quantity of fruit and vegetable consumption or at one specific component of fruits and vegetables might not fully capture the mechanism(s) responsible for a lower cancer risk. Looking at the diversity of fruit and vegetable consumption, reflecting an intake of many different bioactive compounds present in fruit and vegetables, may complement the research on fruit and vegetable consumption and cancer risk. We have previously linked the variety in fruit and vegetable consumption, as measured by diet diversity scores (DDS), to lung and bladder cancer risk in the European Prospective Investigation into Nutrition and Cancer (EPIC). We found no relation between the DDS and bladder cancer risk and a reduced lung cancer risk with increased variety in vegetable consumption. In this current study we evaluated whether the DDS indeed captures a mixed intake of bioactive compounds by comparing the DDS to plasma concentrations of six carotenoids (α-carotene, β-carotene, β-cryptoxanthin, lycopene, lutein and zeaxanthin). Within the cross-sectional study of the EPIC (including 3,089 subjects), we calculated four different DDS based on dietary questionnaires: one represents diversity in vegetable and fruit consumption, two represent diversity in vegetable consumption and one represents diversity in fruit consumption. Plasma carotenoids samples (200 μl) were analyzed by reversed-phase high-performance liquid chromatography (HPLC; HPLC-1100 system, Hewlett Packard). To standardize the six plasma carotenoids, concentrations were re-scaled using a rank difference method; higher values represent larger and lower values represent smaller intra-individual variation in the concentrations of the six plasma carotenoids. Conceptually, the rank difference score should be inversely correlated with the DDS. Correlation coefficients between the DDS and the rank difference score were calculated for the full study population and stratified by subgroups of potential effect modifiers. Complete data on the DDS and plasma levels of carotenoids were available for 2,675 participants. The correlation coefficient (r) between the diversity in vegetable and fruit consumption and the rank difference score was −0.13. Diversity in vegetable consumption was not correlated to the rank difference score. The strongest, although still weak, correlation was found between the diversity in fruit consumption and the rank difference score (r=-0.19). Correlations were somewhat stronger for fasting samples, in women, and in Southern Europe. The diet diversity scores show only weak correlations with rescaled plasma levels of six carotenoids. This might suggest that either questionnaire information or one measurement of plasma carotenoid levels or both cannot fully capture variety in fruit and vegetable consumption. Citation Information: Cancer Prev Res 2011;4(10 Suppl):B7.