Recent evidence suggests a major role for prothrombin as a risk factor for thrombosis. However, estimating prothrombin levels from a deficient plasma-based clotting assay (factor IIc) is expensive and technically difficult in the setting of population-based research. We report the development of an enzyme-linked immunosorbent assay (ELISA) for prothrombin using purified antigen and polyclonal anti-prethrombin-1 IgG. Three different quality control plasmas had coefficients of variation (CV) of 6.5%, 4.9%, and 4.8%. Analytical recovery averaged 103.8%. Results from the ELISA correlated well with factor IIc results (r=0.75). The 5th-95th percentile range for healthy men (n=10) and women (n=16) was 97.7 pg/ml to 161.8 microg/ml. The assay exhibited no significant cross-reactivity with other vitamin-K-dependent proteins. Prothrombin showed no diurnal variation. In a study of biovariability the analytical variability, CV(A), was 3.1%; the within-subject variability, CV(I), was 7.3%; the between-subject variability, CV(G), was 14.5%. The critical difference for sequential values (i.e. the smallest percentage change unlikely to be due to CV(A) or CV(I)) significant at P=0.05 was 21.9%. The index of individuality, CV(I)/CV(G), was 0.50. On the basis of the overall biovariability data, primarily the index of individuality, prothrombin as measured in our ELISA is well suited for applications in population-based research.