Objectives: Previous data have indicated that the erythrocyte membrane may be the preferred sample type for assessing long-chain polyunsaturated fatty acid (LCPUFA) contents in cardiac and cerebral membranes. In this epidemiological study, we examined whether plasma phospholipids can be used for accurate biological monitoring of the LCPUFA state or whether analysis of erythrocyte membrane phospholipids is indispensable. Methods: (1) The analysis of LCPUFA contents in erythrocyte membrane phospholipids was conducted at baseline and after 1 and 3 days at 4°C, and 21 days at −40°C, after blood drawing, and the changes in LCPUFA content were examined. (2) The LCPUFA compositions of plasma and erythrocyte phospholipids in 133 young women (18–30 years old) were examined and the relationships between the sample type and the levels of LCPUFAs were determined. Results: Eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA) and DHA/arachidonic acid (AA) and (EPA+DHA)/AA ratios in erythrocyte membrane phospholipids after 21 days of blood drawing significantly decreased compared with the corresponding baseline data. Regarding AA, EPA and DHA, a significant positive correlation was shown between levels of erythrocyte membrane phospholipids and plasma phospholipids (AA, r=0.364; EPA, r=0.709; DHA, r=0.653). The predictive value of plasma phospholipids for determining the highest concentration quartile in erythrocyte phospholipids was better in EPA (70%) than in DHA (55%) and AA (42%). Conclusions: The measurement of LCPUFA content in erythrocyte membrane phospholipids is necessary for accurate biological monitoring. We also found that LCPUFA in erythrocyte membrane phospholipids is stable in cold storage (4°C) for 3 days after blood drawing.