To compare the relative effectiveness and side effect profiles of antimetabolite drugs in the treatment of noninfectious ocular inflammation. Retrospective cohort study. A total of 257 patients with inflammatory eye disease seen in a single-center, academic practice and treated with an antimetabolite as a first-line immunosuppressive agent from 1984 to 2006. Data recorded included demographics, antimetabolite and prednisone doses, use of other immunosuppressive drugs, response to therapy, and side effects associated with drug use. Ability to control ocular inflammation and to taper prednisone to <or=10 mg daily ("treatment success"); incidence of treatment-related side effects. Ninety patients with inflammatory eye disease were treated with methotrexate, 38 patients were treated with azathioprine, and 129 patients were treated with mycophenolate. Uveitis accounted for the majority of the diagnoses (67%, 66%, and 68% for methotrexate, azathioprine, and mycophenolate, respectively), followed by scleritis (23%, 18%, 17% for methotrexate, azathioprine, and mycophenolate, respectively). The median time to treatment success was 4.0, 4.8, and 6.5 months for the mycophenolate, azathioprine, and methotrexate treatment groups, respectively (P = 0.02, log-rank test). The incidence of side effects was higher in the azathioprine group (0.29/person-year PY) compared with patients treated with methotrexate (0.14/PY) and mycophenolate (0.18/PY). More patients discontinued the drug because of side effects in the azathioprine group (0.24/PY vs 0.09/PY for the methotrexate group and 0.09/PY for the mycophenolate mofetil group). These data suggest that the time to control of ocular inflammation is faster with mycophenolate than with methotrexate. Azathioprine therapy has a higher rate of treatment-related side effects compared with the other 2 agents.