Previous research demonstrates that patients of color are significantly less likely than white patients to receive total knee replacement surgery (TKR), a generally effective and safe treatment for end-stage knee osteoarthritis. These disparities have been found for adults of all ages and of racial/ethnic backgrounds including Black, Latinx, Asian-American/Pacific Islander, multiracial, and First Nations patients (although research has concentrated on Black-white disparities and seniors), and persist even when controlling for insurance type and income. This study investigates utilization rates of TKR in Washington State in 2013 in four different geographic area types (rural low-income, rural higher-income, urban low-income, and urban higher-income), comparing rates for each group by calculating odds ratios. In order to focus on the role of race in physician-patient interactions, odds ratios will be adjusted with logistic regression for age, gender, overall “comorbidity burden” as calculated by the Deyo index, insurance type, and median income in the patient’s zip code. Data for the study comes from the 2013 Washington State Inpatient database, which is derived from hospital billing records and includes roughly 85% of all hospitals in the state. It is important to note that hospitals' methods of demographic information collection are frequently inaccurate, and findings for First Nations patients in particular cannot be interpreted with confidence. Regardless, Black, Latinx, Asian-American/Pacific Islander, multiracial, and First Nations patients are all expected to have lower odds of receiving TKR compared to white patients, a finding which will strengthen the growing conclusion that racial disparities exist independently of access factors and thus that physician-patient interactions likely play a role. No prediction about differences in racial disparities by geographic area type is offered, but these results may lead to new hypotheses about the pathways of the disparities.