November 15, 2007 / Volume 110 / Number 10
Classification of American Indian and Alaska Native Race Race generally is documented by cancer registries based on the patient’s medical record; however, it has been determined that many AI/AN were misclas- sified as non-Native in some central cancer regis- tries21–26 and in state death records.27 Such misclassification leads to underestimation of the true burden of cancer in AI/AN populations.
To improve race classification for AI/AN cases, all records for cases diagnosed from 1995 through 2004 were linked with the IHS patient registration database to identify AI/AN cases that were misclassi- fied as non-Native. The IHS provides medical ser- vices to AI/AN persons who are members of federally recognized tribes. Membership criteria are deter- mined by individual tribes and typically refer to a percentage of ‘‘blood quantum’’ or fraction of AI/AN ancestry, with some tribes requiring only proof of descent from an AI/AN ancestor, whereas others may require as much as one-half ancestry (one-half blood quantum).28 The patient registration database includes persons who have received services since 1985 provided either in an IHS and/or tribal facility or by a non-IHS or nontribal program and paid by IHS Contract Health Services funds. A small number of non-AI/AN persons are eligible for IHS services through marriage or other basis; these records were removed before linkage with the registries. Linkages were conducted using LinkPlus, a probabilistic link- age software program that was developed by CDC that was applied to key patient identifiers (Social Se- curity number, first name, last name, middle initial, date of birth, and date of death).29 Possible matches, requiring manual review, were examined independ- ently by 2 reviewers and, when necessary, were adju- dicated by a third reviewer.
The cancer incidence rates in this report use in- formation from the IHS linkage in combination with information from the race variables that are reported routinely by the central cancer registries. Beginning with cancer cases diagnosed on January 1, 2000 and later, registries report data in multiple race fields for multiracial individuals if that information is available from medical records.30 Coding rules specify that, for persons of multiple races, a nonwhite race takes pri- ority over white race for analytic purposes.31 For this report, all cases classified as AI/AN in the first race field were retained in that category. In addition, when the first race field was classified as white or unknown and there was a positive IHS link, then the case also was reclassified as AI/AN for this report. If the first race field was coded as API, black, or other race and there was a positive IHS link, then the value for first race was retained.
Data for American Indians and Alaska Natives For the AI/AN population in this report, emphasis is placed on data from IHS CHSDA counties, which, in general, contain federally recognized tribal lands or are adjacent to tribal lands (Fig. 1). Unpublished data indicate less race misclassification for AI/AN in these counties than in non-CHSDA counties. Also, the proportions of AI/AN in relation to total population are higher in CHSDA counties than in non-CHSDA counties; 57% of the U.S. AI/AN population resides in the 624 counties designated as CHSDA (Table 1 Supplement A, which shows the AI/AN population by region and residence in CHSDA county, is available at URL: www.seer.cancer.gov/report_to_nation/). Analyses restricted to CHSDA counties are presented for cancer incidence and death rates and stage distri- bution in this report to improve accuracy in inter- preting cancer statistics for AI/AN persons (for distribution of CHSDA counties, see Fig. 1). All results described in the text for AI/AN refer to AI/AN persons who reside in CHSDA counties unless noted otherwise.
The analysis of AI/AN data by IHS region (Alaska, Pacific Coast, Northern Plains, Southern Plains, Southwest, and East) (Fig. 1) was consistent with known regional trends for specific health out- comes and risk factors for AI/AN. Regional analyses have been presented in several publications focusing on AI/AN,32–35 and this approach was preferable to the use of smaller jurisdictions, such as the Adminis- trative Areas defined by IHS,36 which yielded less stable estimates. Supplemental information regarding population coverage and other measures of the bur- den of cancer in the AI/AN by IHS region is con- tained in a Table on the NCI web site (available at URL: www.seer.cancer.gov/report_to_nation/). Addi- tional data regarding cancer incidence and mortality are available from the following sites: www.cancer.- gov (NCI), www.cancer.org (ACS), www.cdc.gov/can- cer/npcr/index.htm and www.cdc.gov/nchs/about/ major/dvs/mortdata.htm (CDC), and www.naaccr. org/CINAP?index.htm (NAACCR).
Socioeconomic Status, Healthcare Access, Risk Factors, Screening, and Stage Data regarding socioeconomic status, healthcare access, behavioral risk factors, and cancer screening by race/ethnicity and geographic region were ob- tained from the U.S. Census Bureau and the CDC’s Behavioral Risk Factor Surveillance System (BRFSS) and were categorized according to Healthy People 2010 objectives.37 Questions concerning cigarette smoking, physical activity, alcohol consumption, and Papanicolaou (Pap) tests (women only) were asked of