Racial misclassification of Native Americans in a surveillance, epidemiology, and end results cancer registry.

BACKGROUND: The cancer incidence for all sites has been reported to be lower in Native Americans than in White Americans. Concerns have been expressed, however, that the observed low incidence may be a result of inaccurate reporting of race.
PURPOSE: The objective of this study was to investigate the extent to which racial misclassification may contribute to the observed low cancer incidence among Native Americans.
METHODS: A registry of individuals eligible to receive medical services funded by the Indian Health Service was linked by computer to the Puget Sound Surveillance, Epidemiology, and End Results (SEER) cancer registry.
RESULTS: Only 137 (60%) of the patients with invasive cancer registered with the Indian Health Service and for whom race was recorded were identified as Native Americans in the SEER registry. Similarly, 55 (69%) of 80 in situ cervical cancer case patients were classified as Native American. A strong association was observed between Native-American blood quantum level and racial misclassification.
CONCLUSION: The results of this study indicate that the observed low cancer incidence in Native Americans relative to Whites in the northwest United States is at least partially attributable to racial misclassification in the SEER cancer registry.