Few studies have adequately considered the type of seafood and background dietary factors when evaluating diet-biomarker and diet-disease associations. The objective of this paper is to evaluate the relationship between different seafood meals and long-chain (n-3) fatty acids [eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA)] biomarkers in the Multi-Ethnic Study of Atherosclerosis (MESA) with white, Chinese-American, black, and Hispanic participants. Dietary intake from a FFQ and plasma phospholipid fatty acids were assessed in 900 MESA participants who were not taking fish oil supplements. When simultaneously adjusting for all seafood groups, concentrations of EPA and DHA in plasma phospholipids were positively correlated with nonfried fish consumption in all 4 ethnic groups (r = 0.24-0.46; P < 0.01) but not with nonfried shellfish, fried fish, or fish in mixed dishes. The magnitude of this correlation was attenuated by up to 67% when type of seafood was not taken into account. After further adjusting for demographic characteristics and other dietary characteristics in multivariate regression models, the association of nonfried fish consumption remained significant (P-trend < 0.001). Data were suggestive of a plateau effect at a nonfried fish intake of about twice weekly. The association of nonfried fish consumption was not modified by intake of (n-6) PUFA or alpha-linolenic acid. This study highlights the importance of cooking methods (nonfried vs. fried fish), types of seafood (fish vs. shellfish), and the overall seafood consumption when assessing health effects of long-chain (n-3) fatty acids of seafood consumption.