OBJECTIVES:   To investigate the effect of Medicare’s cut in chemotherapy reimbursement (implemented in 2004-2005) on rituximab prescription among Non-Hodgkin’s Lymphoma (NHL) patients and to estimate the marginal survival benefits of rituximab using an instrumental variable (IV) approach. METHODS: Using the Surveillance, Epidemiology and End Results (SEER)-Medicare linked database, we studied patients 66 years or older diagnosed with first primary NHL from 2002-2007 (N=19,424).  Logistic regression was used to assess the impact of Medicare reimbursement cut on first course rituximab prescription, after controlling for clinical and non-clinical factors.  Two stage least square regressions using Medicare reimbursement policy as IV were used to estimate whether the changes in the rituximab prescription rates could lead to one and three years of survival benefit.  RESULTS:   In the post-policy period (2004-2008) the likelihood of rituximab prescriptions increased as compared to the pre-policy period (AOR of 1.52 in 2004, AOR of 1.79 in 2005 and after, respectively).  Using IV approach, we found that the first course of rituximab prescription among NHL patients significantly improved one- and three-year survival (effectiveness estimate=0.37 for one-year survival, effectiveness estimate=0.44 for three-year survival respectively). CONCLUSIONS:   Rituximab prescriptions increased in response to the Medicare’s reimbursement cut.  Substantial rituximab survival benefits were observed among marginal NHL patients whose treatment choices were affected by the policy change.