Unique circumstances of culture, location, history, and healthcare combine to produce unusual patterns of cancer occurrence among American Indians and Alaska Natives (AI/AN) in the United States.1 Although there is great diversity among the hundreds of tribes and linguistic groups that comprise the AI/AN population, they share many features, including a history of defeat, displacement, and cultural trauma in the not-so-distant past, as well as a rich cultural heritage that can be a source of strength in the present. Many AI/AN live on reservation lands or in remote rural areas, with primary healthcare needs provided only by the Indian Health Service (IHS) or tribally operated health programs.2, 3 Both rural and urban AI/AN populations experience greater poverty, lower levels of education, and poorer housing conditions than the general US population.4 The risk factor profile for many AI/AN communities is quite different from others, including use of tobacco and alcohol, physical activity level, and obesity.5, 6 Dietary practices and exposure to harmful environmental agents are less well documented, yet may also be important risk factors. Many of these influences on the burden of cancer for this group tend to be lost when data from AI/AN cancer cases from all regions of the United States are combined into composite rates. Careful study of the incidence of cancer among AI/AN can be helpful in understanding the influence of different exposures, in addition to quantifying the burden of disease for this population. The special collection of papers in this Supplement uses combined cancer incidence data from the 2 major US cancer registry systems, enhanced by record linkages and geographic factors, to improve racial classification.1, 6–17 The 2004 diagnosis year represents the first time that population-based cancer incidence data for AI/AN have been available regionally and nationwide for analysis.1, 6–17
The collection of papers in this Supplement combines cancer incidence data from the National Program of Cancer Registries and the Surveillance, Epidemiology, and End Results program, enhanced by record linkages and geographic factors, to provide a comprehensive description of the cancer burden in the American Indian/Alaska Native population in the United States. Cancer incidence rates among this population varied widely, sometimes more than 5-fold, by geographic region. Cancer 2008;113(5 supp):1113–6. Published 2008 by the American Cancer Society.
The unusual risk factor profile of AI/AN communities is the topic of 1 of the papers in this collection.6 American Indians have used tobacco for many generations, and it is considered by many to have medicinal and spiritual properties. Historically, use of tobacco was limited to small quantities and specific ceremonial settings. Since WWII, however, the habitual use of commercial tobacco products has greatly increased, and AI/AN now have the highest smoking prevalence in the country, over 50% in some communities.5, 6, 19 A notable exception is the tribes of the Southwest, who have remarkably low prevalence of tobacco use.
For a surprising number of AI/AN, traditional food sources from farming, hunting, and fishing make up a significant portion of their diet, which may serve to protect against the possible harmful effects of a modern Western diet but leaves them vulnerable to higher levels of contaminants, such as heavy metals, that are concentrated in wild food sources.
Sources of Healthcare
AI/AN who are enrolled members of federally recognized tribes are eligible to receive, without charge, primary healthcare from the IHS system of Federal and Tribal hospitals and clinics. Those who do not live close to an IHS facility may need to travel great distances to receive even basic primary care. Cancer treatment, like most specialty care, is purchased from non-Federal providers through the IHS Contract Health Service, which is available only to IHS beneficiaries who live in certain US counties (the Contract Health Service Delivery Area, or CHSDA). Restrictions on eligibility for the Contract Health Service mean that modern cancer treatment for many is logistically difficult and unaffordable.
Cancer Incidence Data
The entire Native population is only about 1.1% of the US population,20 and they are dispersed throughout the country, making it difficult to collect aggregate data on cancer incidence in AI/AN populations. Most previous publications have relied on mortality data from death certificates21 or have reported on incidence in limited areas covered by state or local cancer surveillance systems, notably those in the Surveillance, Epidemiology, and End Results (SEER) cancer registry system.22‒26
The National Cancer Institute (NCI) has funded and supported the SEER Program since 1973.27 SEER routinely collects data on patient demographics, primary tumor site, tumor morphology and stage at diagnosis, first course of treatment, and follow-up for vital status. Recognizing the importance of understanding cancer patterns in minority populations, NCI provides data for American Indians in Arizona and New Mexico and Alaska Natives in the SEER system, and more recently has begun to develop tribally based cancer registries at the Cherokee Nation of Oklahoma and in the Northwest. SEER registries currently cover 26% of the total US population, including 42% of AI/AN. The cancer data from these sources has been intriguing, showing that Alaska Natives have rates of lung, colon, and breast cancer 5 times or more higher than those of Southwestern Indians, whereas rates of stomach, kidney, uterine cervix, and liver cancer are similar in the 2 regions, and higher than in non-Hispanic whites (NHW).
Studies using death certificates21 and IHS hospital data28 indicated that cancer rates in the Northern Plains and other regions were as high as in Alaska. However, population-based cancer incidence data were not available for American Indians outside of the Southwest until recently.
Established by Congress through the Cancer Registries Amendment Act in 1992, and supported by the Centers for Disease Control and Prevention (CDC), the National Program of Cancer Registries (NPCR) collects data on the occurrence of cancer, including demographics, tumor characteristics, stage of disease at diagnosis, and type of initial treatment.29, 30 Today, NPCR supports central cancer registries in 45 states, the District of Columbia, Puerto Rico, the Republic of Palau, and the Virgin Islands. NPCR data are used by states to monitor the cancer burden, identify cancer incidence variation for racial and ethnic populations, provide data for research, provide guidance for health resource allocation, respond to public concerns and inquiries about cancer, and evaluate cancer prevention and control activities. NPCR currently covers 96% of the US population, including 90% of the AI/AN population.
NPCR and SEER cover 100% of the US population and together represent an opportunity to report on cancer incidence for the AI/AN population nationally and regionally.18 This collection of papers addresses site-specific cancers of interest, risk factors among AI/AN, and methods used by the collaborating authors to improve racial classification in cancer registries.
Correcting Misclassification of Race
In the past, few central cancer registries were concerned about correctly identifying AI/AN in their databases, where race and ethnicity were abstracted from medical records. Numerous studies have demonstrated that many AI/AN persons were misclassified as another race in cancer registry data, and that the extent of the misclassification varied by registry.31‒36 When coupled with population denominators from the US Census Bureau, where race is self-identified, this had the effect of lowering apparent cancer rates for AI/AN. To address this problem, IHS, NCI, CDC, and all NPCR and SEER cancer registries launched a unique collaboration to link IHS beneficiary records with cancer registry databases. Linkages to identify AI/AN cases misclassified as non-Native were conducted using LinkPlus, a probabilistic linkage software program developed by the CDC that identifies records representing the same individual in the IHS and registry databases.37, 38
In addition, CHSDA counties and IHS regions were geographic factors used to further elucidate the burden of cancer incidence in the AI/AN population.37 CHSDA counties, in general, contain federally recognized tribal lands or are adjacent to tribal lands. The proportion of AI/AN relative to the total population is higher in CHSDA counties, with less misclassification than in non-CHSDA counties. The analysis of AI/AN data by IHS region conforms with known regional patterns of specific health outcomes and disease risk factors for AI/AN. These processes greatly improved the identification of race and the description of the cancer burden in NPCR and SEER registries for AI/AN populations in the United States.
Key Findings and Questions
In these reports, cancer incidence rates in AI/AN were lower than rates in NHW for most cancer sites. Notable exceptions included higher rates of cancers occurring in the kidney, stomach, cervix uteri, liver, and gallbladder. Are these differences because of shared risk factors, genetic predisposition, or lack of screening and preventive services? Although rates were significantly declining for most sites among non-Hispanic whites (including the leading cancer sites), the declines among AI/AN were generally smaller and not statistically significant. This raises questions about whether AI/AN people with cancer have access to appropriate and timely screening and treatment.
Cancer incidence rates among AI/AN varied widely (sometimes 5-fold) by IHS region. Lung, colorectal and breast cancers occurred at rates higher or similar to the NHW population in Alaska and the Northern Plains, but were dramatically lower in the Southwest. Lung cancer tracks closely with the smoking rates in those regions, but how can we explain the differences in breast and colon cancer?
The culmination of this collaborative work represents important progress in providing accurate data on the burden of cancer in the AI/AN population nationwide and by region. We hope that this information will be useful to guide public health and clinical policy for cancer control, and to generate hypotheses for etiologic research.