Several variants at DBH are significantly associated with plasma DbetaH activity (pDbetaH). However, the overwhelming majority of data on this genotype-phenotype relationship has been gathered in samples from Europeans and European Americans (EAs). In this study, we examined the relationship between DBH polymorphisms and pDbetaH in samples from African-American (AA) subjects. Genotypes were determined at a 19-bp insertion/deletion polymorphism (ins/del) and four single-nucleotide polymorphisms (SNPs) at DBH in 109 samples. Analyses were performed using analyses of variance (ANOVAs) (for individual SNPs) and regression procedures (to assess the joint effects and the specific SNP-based haplotypes). We found: (1) single-variant analysis of all polymorphisms revealed apparent associations to pDbetaH, with rs1611115 accounting for the largest proportion of the variance in pDbetaH (28.7%) and ins/del the smallest (6.5%); (2) modest but significant linkage disequilibrium (LD) existed between ins/del and rs1611115; (3) LD between all other pairs of variants was not observed; (3) stepwise regression showed that a model containing rs1611115, rs2519152 and rs6271 accounted for 37.6% of the variance in pDbetaH, with rs6271 showing additional 7.6% above the effect of rs1611115, and rs2519152 showing additional 2% above rs1611115 and rs6271; (4) two common haplotypes, C-T-C and T-C-C at rs1611115-rs2519152-rs6271 were significantly associated with pDbetaH (P=0.0025 and 0.0036, respectively). The data support the validity of prior reported associations and underscore the importance of analyzing multiple SNPs across DBH in future association studies examining disease and biochemical phenotypes.