OBJECTIVES: The development of new medical therapies can have a profound effect on the epidemiology of a disease, especially when that development represents the first efficacious treatment. Rituximab was the first monoclonal antibody therapy for cancer, gaining FDA approval in 1997 in the United States (US) to treat non-Hodgkin’s lymphoma (NHL). Rituximab’s positive impact on mortality has been well-documented. The potential effect of rituximab on the incidence of NHL is necessary to understand the full epidemiological impact of the drug. This study aimed to investigate that effect. METHODS:  Age-adjusted incidence rates of NHL were modeled over time using Poisson regression, allowing a different slope for change in yearly rate before and after 1997. This allows assessment of whether the incidence rate was accelerating, decelerating, or did not change after 1997. To determine the optimal change-point for incidence of NHL, the model was repeated for each year within the study period and Akaike information criteria (AIC) were compared. The year with the lowest AIC was considered the optimal change-point. This analysis utilized the Surveillance, Epidemiology, and End Results (SEER) 9 incidence data, which covers the time period 1973–2010 (accessed using SEER*Stat software, version 8.1.2). RESULTS:  From 1973–2010, there were 239,118 total cases of NHL documented in SEER. The model estimated an increase in incidence of 1.9% per year prior to 1997, compared to an estimated increase of only 0.5% per year from 1997–2010. Comparison of AIC’s identified 1990 as the optimal change-point (annual incidence increase pre-1990: 2.4%; post-1990: 0.8%). CONCLUSIONS:   The incidence of NHL in the US has been increasing fairly steadily since 1973, though it has been increasing more slowly in recent years, and nearly stabilized after 1997. In fact, this deceleration appears to have started around 1990, prior to the approval and availability of rituximab.