e19023 Background: The use of EGFR monoclonal antibodies extends life for patients with advanced colorectal cancers (CRC) whose tumors exhibit wild type KRAS. Research suggests that KRAS testing is underused, denying patients access to targeted therapy. We sought to study the role of socioeconomic factors in the application of KRAS testing. Methods: We identified subjects with stage IV colorectal adenocarcinoma diagnosed between 2010 and 2015 in the Surveillance Epidemiology and End Results (SEER) database. We identified clinical and demographic characteristics, including diagnosis year, race, ethnicity, sex, age, geography, median household income (MHI), insurance coverage, tumor grade, and tumor location. We used multivariable logistic regression models to evaluate associations between these factors and the rate of KRAS testing. We constructed Cox proportional hazard models to assess the impact of KRAS testing on survival. Results: We identified 37,676 patients with stage IV CRC, 31.1% of whom were tested for KRAS mutations. The rate of KRAS testing varied from 23.2% to 56.5% across geographic regions. Patients were more likely to be tested if they were younger (OR 5.10 for age 20-29 vs 80+, 95% CI 3.99-6.54, p < 0.0001), diagnosed more recently (OR 1.92 for 2015 vs 2010, 95% CI 1.77-2.08, p < 0.0001), or lived in an area of high MHI (OR 1.24 for MHI of > $69,311 vs < $49,265, 95% CI 1.14-1.35, p < 0.0001). Patients were less likely to be tested if they had Medicaid (OR 0.83, 95% CI 0.77-0.88, p < 0.0001), were unmarried (OR 0.78, 95% CI 0.75-0.82, p < 0.0001), or had rectal cancer (OR 0.79, 95% CI 0.74-0.84, p < 0.0001). The risk of death was decreased in patients who received KRAS testing (HR 0.77, 95% CI 0.75-0.80, p < 0.0001). The median survival for patients who were tested was 20 months (95% CI 20-21), vs 11 months (95% CI 10-11) for patients not tested (p < 0.001). Conclusions: We found a low rate of KRAS testing overall in CRC patients diagnosed between 2010 and 2015. Testing rates increased over the study period, but, even as of 2015, the majority of patients did not have KRAS testing.We found thatpatients with Medicaid and those who lived in areas of lowest income were less likely to have KRAS testing. Prior studies reported that Medicaid was associated with lower utilization of KRAS testing, but ours represents a more contemporary cohort and is the first to evaluate Medicaid and MHI simultaneously. Ideally Medicaid insurance would increase access to molecular testing for patients with limited resources, but our model suggests that income-based disparities persist despite Medicaid insurance.