Variability among individuals in their therapeutic response to psychotropic drugs and in susceptibility to adverse effects is considerable. Pharmacogenetics addresses the contribution of genetic factors to this variability. An important focus of interest in pharmacogenetics has been on candidate genes that play a role in susceptibility to the antipsychotic drug-induced adverse effect, tardive dyskinesia (TD). Four published studies have reported an association between a serine (ser) to glycine (gly) polymorphism in exon 1 of the dopamine D3 receptor gene (DRD3) and TD; three failed to replicate this finding and one found an insignificant trend. We examined the association in a pooled sample of 780 patients (317 with TD and 463 without TD) drawn from 6 research centers, who were divided into 8 groups based on their population origin. The analysis employed stepwise logistic regression so as to allow confounding effects of group, age, and gender to be taken into account. TD was significantly associated with DRD3 gly allele carrier status (x(2)=4.46, df 1, p =.04) and with DRD3 genotype (x(2)=6.62, df 2, p =.04) over and above the effect of group. Similar positive effects were observed when controlling for age and gender (x(2)=5.02, df 1, p =.02 for gly allele carrier status; x(2) = 7.51, df 2, p =.002 for genotype). Examining abnormal involuntary movement scores as a continuous variable, we found that patients homozygous for the gly allele had significantly higher scores than ser-gly heterozygotes (p =.006) or ser-ser homozygotes (p <.0001). We also performed a meta-analysis that included, besides the groups in the combined analysis, three other published studies on DRD3 and TD. The Mantel-Haenszel pooled odds ratio for DRD3 gly allele carrier status increasing susceptibility to TD was 1.33 (95% CI 1.04-1.70, p =.02); the cumulative pooled estimate showed an odds ratio of 1.52 (95% CI 1.08-1.68, p <.0001). These findings support a small but significant contribution of the DRD3 ser9gly polymorphism to TD susceptibility that is demonstrable over and above population effects and the effect of age and gender on the phenotype.