Mycophenolate mofetil has replaced azathioprine in many transplant centers, but data on hard outcomes such as graft and patient survival are lacking. We analyzed graft and patient survival of the 1219 first renal allograft recipients at the Medical University of Vienna who were treated with azathioprine or mycophenolate mofetil over the past decade. Cox's proportional hazards models were used to compute crude and confounder-adjusted hazard ratio estimates of azathioprine versus mycophenolate mofetil. Model building was performed by applying the purposeful selection algorithm, confounding by indication was addressed by propensity scores and marginal structural models. Five years after transplantation, 12% of mycophenolate mofetil users experienced functional graft loss whereas 26% of the azathioprine users had lost their graft (P<0.001). The hazard ratio for functional graft loss was 2.15 (95% confidence interval 1.16-4.02, P=0.016) in azathioprine versus mycophenolate mofetil patients. Actual graft loss at five years had occurred in 25% of the mycophenolate mofetil patients and 49% of the azathioprine users (P<0.001); hazard ratio 2.04 (95% confidence interval 1.22-3.39, P=0.006). Patient survival was not different in any of the analyses. The data from our observational study suggest that mycophenolate mofetil use was associated with a lower risk of graft loss than azathioprine-based immunosuppression.