Studies by independent researchers who used data from agencies, hospitals, and state health departments provide additional insight into the cancer burden among members of ethnic minority and medically underserved groups. In many cases, as discussed below, these studies highlight important differences between regional data and national estimates provided by the SEER program.
A study that used hospital discharge data from IHS during the period between 1980 and 1987 provides estimates of cancer incidence and mortality for the 11 IHS regions (Nutting et al., 1993). A high degree of variation in the occurrence of cancer was seen among Native Americans in different geographic regions (see Tables 2-16 and 2-17). For example, the incidence of cancer among American Indian males in the Billings, Montana, area was nine times higher than that among American Indian males of the Phoenix area (56.4 per 100,000 population, in contrast to 6.3 per 100,000 population). Female breast cancer incidence also varied widely according to geographic region. The incidence in the Phoenix area was 18.8 per 100,000 population, in comparison to 51.5 per 100,000 population among Alaska Natives.
The IHS data for cancer incidence among American Indian women also illustrated large regional differences not shown by the SEER program estimates. For example, the incidence of cancer of the cervix in the Albuquerque, New Mexico area was approximately two- to threefold higher than seen in Portland, Bemidji, and Oklahoma.
A second study was initiated to determine the extent of racial misclassification of American Indians in the Washington State Cancer Registry when compared to data obtained from the Portland Area Indian Health Service and tribal specific cancer data (Sugarman et al., 1996). The estimated age-adjusted cancer incidence among American Indians in Washington state increased from 153.5 per 100,000 population before record linkage to 267.5 per 100,000 after linkage. This study found that most cases of cancer in American Indian individuals were identified by the IHS registry and a few additional cases were identified using the tribal rolls. More than one quarter of American Indians classified as full heritage (100 percent blood quantum) were not coded as American Indians in the tumor registry, again suggesting that true misclassification frequently occurs.
Special studies that use data from the SEER program are also a source of important information regarding the incidence of cancer among members