Background: Recent data have implicated tumor necrosis factor (ligand) superfamily, member 4 (TNFSF4) gene variation in myocardial infarction in women; however, no prospective data are available on either incident arterial or venous disorders., Methods: We evaluated 2 previously characterized TNFSF4 gene variants (-921C>T and dbSNP rs3850641) with a) incident arterial events using a prospective case-cohort design with 344 incident CVD cases and 2254 control participants, all white, drawn from the Women's Health Study cohort with 10 years of follow-up, and b) venous thromboembolism (VTE) risk using a nested, matched case-control design of 108 white male pairs (drawn from the Physicians' Health Study cohort) and a case-cohort design of white female participants consisting of 125 cases and 2269 controls (drawn from the Women's Health Study cohort), analyzed separately., Results: Genotype distributions were in Hardy-Weinberg equilibrium. Results from a marker-by-marker regression analysis, adjusting for traditional risk factors, showed a significant association of -921C>T with an increased risk of VTE in women (additive: odds ratio 1.86; 95% CI 1.17-2.92, P = 0.008) in women. Furthermore, using a haplotype-based regression analysis, haplotype C-G was associated with a reduced risk of VTE relative to the referent haplotype, C-A (odds ratio 0.50; 95% CI 0.27-0.92; P = 0.02). In contrast, we found little evidence for an association of the variants/haplotypes with risk of VTE in men or CVD risk in women (as previously reported)., Conclusions: Our present findings, if corroborated in other prospective investigations, suggest that the TNFSF4 variants tested may be useful indicators for assessing the risk of venous thromboembolism.