The adjusted relative risk (aRR) for development of post-transplant lymphoproliferative disorder (PTLD) is higher in kidney transplant recipients receiving monoclonal antibody induction therapy, but the aRR between the different available polyclonal agents has not been investigated in detail. We analyzed data from the United Network of Organ Sharing registry on all kidney transplants performed between 1987 and 2003. The aRR for PTLD development was calculated using SAS 9.0 statistical software and Cox proportional hazards modeling, adjusting for multiple covariates. There were 539 cases of PTLD among 84 907 kidney transplant recipients, who received either polyclonal antibody induction or no induction therapy. In adjusted analysis, the aRR for PTLD development (vs. no induction) was significantly higher with use of equine anti-thymocytic globulin (E-ATG; aRR = 1.61, p = 0.0003) or anti-lymphocytic globulin (ALG; aRR = 1.35, p = 0.0055) but not with rabbit anti-thymocytic globulin (R-ATG; aRR = 1.17, p = 0.29, NS). Median follow up times were significantly shorter in the R-ATG cohort than the ALG or E-ATG cohort (median 368 vs. 1433 and 2055 day). However, in an analysis restricted to pediatric recipients, where median times to PTLD are less than 200 days, only E-ATG was associated with a higher aRR for PTLD (aRR = 2.16, p = 0.0078), while R-ATG and ALG were not. There is a higher aRR for PTLD after kidney transplantation with E-ATG, but not R-ATG. This may only partially be explained by shorter follow up time and may represent differential hazard for PTLD among the agents.