S599 Racial Disparities in Tumor Location Documentation in Colorectal Cancer: A Population-Based Study

Introduction: Colorectal cancer (CRC) incidence is rising in younger adults, and racial disparities in screening, diagnosis, and outcomes are well documented. Black and other minority patients are less likely to receive timely CRC screening, more likely to be diagnosed at later stages, and experience worse survival rates. Tumor location, particularly whether a tumor is left or right sided, can influence stage at diagnosis, treatment approach, and prognosis. However, it is unclear whether tumor location is documented consistently across racial groups. This study examined patterns of tumor location documentation in CRC patients to assess whether disparities exist in how tumor site is recorded. Methods: Data from the SEER 17 Registries (2000–2022) was analyzed, identifying patients diagnosed with malignant colon cancer using ICD-O-3 codes C18.0–C18.9. Appendix cancers (C18.1) were excluded. Tumors were grouped as left-sided (C18.5–C18.7), right-sided (C18.0–C18.4), or not otherwise specified/overlapping (C18.8–C18.9). Patients were stratified by race (White, Black, Asian/Pacific Islander, American Indian/Alaska Native, Unknown). SEER*Stat frequency sessions were used to calculate the proportion of cases with tumor location documented and the proportion unspecified. Among cases with a documented tumor location, we further examined the differences between locations, categorized by left sided, right sided, or not otherwise specified (NOS) coding. Results: Among 1,147,456 CRC cases, 59.8% of White patients had tumor location documented, compared to 53.6% of Black patients, 56.4% of Asian/Pacific Islander patients, and 52.8% of American Indian/Alaska Native patients. The “Unknown” group had the highest unspecified rate (61.4%). Even among White patients, only 59.8% had location documented, well below expected clinical standards. Among cases with tumor side recorded, the proportion labeled as “NOS” was similar across races, ranging from 49.7% to 51.8%. A chi-square test confirmed that differences in documentation rates by race were statistically significant (Pearson χ² = 33,805.0, df = 4, P < 0.001). Conclusion: Substantial proportions of CRC cases, particularly among non-White patients, lack clear documentation of tumor location. Even when recorded, nearly half are coded as NOS, limiting clinical utility. These findings suggest that disparities in CRC include not only outcomes, but also data quality. Improving documentation is essential for equity in CRC care and research.