The effect of anxiety on dietary intake of humans has been investigated through a number of laboratory, clinical and cross-sectional studies; no prior study, however, has examined potential associations between anxiety and overall dietary patterns. Aim of the present work was to describe dietary patterns in relation to anxiety trait in a nationally representative sample of Greek adults from the ATTICA Study. A sample of 453 men and 400 women were randomly selected from various areas of Attica region, Greece. Anxiety levels were assessed through Spielberger State-Trait Anxiety Inventory. Dietary habits, socio-demographic and lifestyle characteristics were recorded for all participants. Principal component analysis was used for the extraction of dietary patterns. More anxious, compared to less anxious, men and women exhibited different dietary patterns. In particular, the "light" dietary patterns that were emerged in the less anxious men and women did not appear as distinct patterns among men and women in the upper anxiety tertile. In women, a "Western-type" diet explained two times greater variance of food intake of those in the upper-anxiety tertile, compared to their counterparts in the low tertile. A vegetarian pattern was found only among the less anxious women, who also exhibited the lowest consumption of red meat and sweets. Regression analysis supported and further elucidated previous results: after adjusting for potential confounders, sweets intake, as well as meat and products intake, were positively associated with anxiety score in females; in males a negative association was found with legumes/cereals intake. From a public health point of view, given the increased prevalence of anxiety and other mental disorders, these findings should be taken into account when designing and evaluating interventions for the general population.