People in circumpolar regions of Canada are experiencing rapid changes in their social, cultural, and physical environments. Scientists are increasingly concerned about the effects of these changes on population health, particularly the health of Aboriginal populations, which includes First Nations, Inuit, and Métis groups. The increasing prevalence of "diseases of modernization," such as diabetes and high blood pressure, is already being observed (1). In northern Canada, more than half of the population is composed of Aboriginal people, primarily First Nations and Inuit, unlike most other circumpolar regions where Aboriginal people are a minority (2).

Although northern Canada is sparsely populated and communities are geographically disparate, this region has experienced rapid growth in some areas in recent years. The total population recently exceeded 100,000 (3); the largest increase occurred in the Northwest Territories (NWT) and in Nunavut. The Canadian north enjoys a higher per capita gross domestic product (GDP) than Canada as a whole, and the GDP of the NWT is about 2.5 times that of Canada's GDP, primarily attributable to increased activity associated with the extraction of natural resources (4). However, this wealth is not evenly distributed; socioeconomic indicators show that Aboriginal people in the region are disadvantaged (2).

Across Canada, Aboriginal populations experience substantial variability in living conditions and lifestyles. Some illness and death patterns at the population level are drawing increased attention. For example, the prevalence of some chronic diseases, including diabetes, heart disease, and high blood pressure, is higher among the First Nations population than among the rest of the Canadian population; the prevalence of some risk factors, such as smoking and infrequent physical activity, is also substantially higher among the First Nations population than among the rest of the Canadian population (5). Although the prevalence of many chronic diseases is lower among the Inuit than among the rest of the Canadian population, the prevalence of smoking is higher and obesity is also increasing (2). However, not enough research has been conducted that explores the variation of these chronic diseases and their risk factors within the populations of northern Canada.

There were 2,074 respondents from northern Canada in cycle 1.1 and 2,166 respondents in cycle 3.1. Approximately one-quarter (26.7%) of respondents in cycle 1.1 were from Nunavut, one-third (33.6%) were from Yukon, and the remainder were from NWT (39.8%) (Table 1). The geographic distribution of respondents was similar in cycle 3.1. Approximately 40% of respondents in each cycle were of Aboriginal ethnicity, although ethnic composition varied across the 3 regions.

In the first cycle, the percentage of Aboriginal respondents who reported having any chronic health condition was lower compared with the percentage of non-Aboriginal respondents (Table 2). However, there was a large increase between the 2 cycles for Aboriginal respondents (8.8 percentage point difference), but not for non-Aboriginal respondents (0.6 percentage point difference). This increase was observed for both younger (20-54 years) and older (≥55 years) age groups (Figure). In the first cycle, Yukon respondents reported the highest crude prevalence of any chronic condition, and prevalence estimates were similar for NWT and Nunavut respondents. Prevalence estimates for any chronic condition were similar for all regions between cycles 1.1 and 3.1, with the exception of NWT, where an increase was observed.

The prevalence of all of the investigated chronic diseases was similar in cycles 1.1 and 3.1. Arthritis was the most common chronic condition reported; it was reported by 13% of northern Canada respondents, followed by high blood pressure (approximately 10%) and asthma (approximately 8%) ( Table 2). Prevalence estimates were slightly higher for respondents from Yukon than for respondents from the other 2 regions. Data for some diseases and risk factors were not included in our analyses because of low frequencies. Specifically, data for bowel disorders, cancer, emphysema/COPD, and stroke were not reported.

Approximately one-quarter of Aboriginal and non-Aboriginal respondents were classified as overweight, and approximately 20% were classified as obese in Cycle 1.1 (Table 2). Between the 2 cycles of data collection, there was an increase (5.3 percentage point difference) in overweight for non-Aboriginal respondents, but not for Aboriginal respondents. There was an increase in obesity for Aboriginal (5.2 percentage point difference) and non-Aboriginal (2.6 percentage point difference) respondents. Approximately one-third of NWT respondents were classified as overweight in cycle 1.1; prevalence was lower in the other 2 regions.

More than one-half of Aboriginal respondents were identified as daily smokers in both cycle 1.1 (52.5%) and cycle 3.1 (50.2%), compared with less than one-third (29.9%) of non-Aboriginal respondents in the first cycle and less than one-quarter (23.5%) of the latter group in the second cycle. Daily smoking rates were highest in Nunavut (50.2%) and lowest in Yukon (27.8%) in cycle 3.1. Regular drinking and binge drinking were more common among non-Aboriginal and Yukon respondents. Higher prevalence of infrequent physical activity and inactivity during leisure time was observed among non-Aboriginal and Nunavut respondents (Table 2).

The adjusted odds of having any chronic health condition were significantly lower and the odds of heart disease were significantly higher for Aboriginal respondents in cycle 1.1, but the AORs for these 2 variables were not significant (any chronic condition, P = .61; heart disease, P = .78) in cycle 3.1 (Table 3). Among Aboriginal respondents, in both survey cycles, odds of daily smoking and infrequent physical activity were higher than for non-Aboriginal respondents, but odds of regular drinking and binge drinking were lower. Moreover, the ORs for daily smoking and infrequent physical activity were larger in cycle 3.1 than in cycle 1.1.

By region, the odds of having any chronic condition were highest for Yukon respondents in cycle 1.1 (Table 3). The odds of arthritis were lowest for Nunavut respondents in cycle 1.1. The odds of asthma were lower for Nunavut respondents in both cycles. There were no regional differences for other chronic diseases.

For Yukon respondents, although there were a number of significant differences in the AORs for risk factors, those for obesity and inactivity during leisure time (lower odds) were observed in both cycles (Table 3). For Nunavut respondents, the odds of daily smoking were higher than for respondents from the other regions in both cycles, and the odds of regular drinking were significantly lower in cycle 1.1.

These population-based health survey data suggest that chronic diseases and risk factors are not uniformly distributed across ethnic groups and geographic regions of northern Canada and that some negative shifts have occurred in their distribution in a relatively short period of time.

Some of the findings of this study are consistent with previous research comparing Aboriginal and non-Aboriginal populations in southern Canada or in other countries, but others are not. For example, smoking rates are reported elsewhere to be higher among Aboriginals compared with non-Aboriginals (8,12). A previous study that compared chronic disease and risk factors between northern and southern Canadian populations found that the Aboriginals had a higher prevalence of overweight and obesity, daily smoking, and physical inactivity than the non-Aboriginals (13). However, we found that these risk factors among northern Canadian populations varied substantially. In addition, with respect to overweight, recent studies suggest that Canadian Aboriginals have substantially higher rates of overweight than non-Aboriginals with self-reported white ethnicity (14,15), which contradicts our findings. The difference may arise, in part, because the non-Aboriginal group in this study was not limited to whites.

Overweight, obesity, smoking, binge drinking, and inactivity threaten the long-term health of northern Canadian populations. Excess weight and smoking increase the risk of cardiovascular disease (16), smoking is associated with an increase in certain forms of cancer, and excess weight can also contribute to the development of chronic musculoskeletal conditions.

A recent study, which reviewed the literature on self-reported cancer screening behaviors and race, found greater bias in self-reported behaviors for racial and ethnic minority groups, including North American Indians; it suggested that a variety of factors, including motivation, cognition (ie, health literacy and knowledge of screening techniques), and the health care environment, may contribute to bias in estimates from self-reported data (17). Chronic disease prevalence estimates may not agree with estimates obtained from other population-based sources. For example, 1 study found that agreement between survey and administrative data was highest for diabetes and high blood pressure, but much lower for arthritis and heart disease (18).

There are several limitations to our study. We relied on self-reported data, which may have underestimated the prevalence of overweight, obesity, smoking, and binge drinking and overestimated the prevalence of physical activity (19). Health care services and screening programs for Aboriginal people in northern Canada may be less accessible than for non-Aboriginal people; this may result in biased self-reporting about chronic diseases because the survey questions in this study focused on conditions diagnosed by a health professional (8). Measures of overweight and obesity developed in European populations may not be appropriate for Aboriginal populations (20), especially the Inuit, whose ancestors came from Asia (21); in general, people of Asian descent have more risk factors and higher body fat than people of European descent with the same BMI. Leisure time physical activity does not include activities performed at work (eg, work-related labor activity), at school (eg, physical education class), or at home (eg, household chores). The measure of any chronic condition used in this study provides an overall indicator of health status but obscures changes in individual conditions; an increase in this indicator does not necessarily represent an increase in the prevalence of all chronic conditions. The CCHS covers 90% of the Yukon population and 97% of the NWT population, but only 71% of the Nunavut population (22) (data were only collected in the 10 largest communities in Nunavut). Thus, the results for the latter region may not be as representative of the entire population as results for the other 2 regions. Finally, we used cross-sectional data to estimate change in chronic diseases and risk factors instead of using repeated measurements on the same cohort of respondents to study the progression of health.

Although these survey results can provide snapshots of health status and its determinants, further analyses and other population-based data sources are needed to obtain a more complete picture of chronic disease in northern Canada, its association with health care use, and the effect of community-based interventions on changes in health status. Not enough information exists about different Aboriginal groups, particularly Inuit and First Nations groups, who do not have identical disease burden or risk factor distribution. Data from multiple cycles of the CCHS may be combined to achieve sufficient sample size to enable comparisons of First Nations and Inuit groups within northern Canada. Administrative data, including hospital and physician records, are 1 source of routinely collected data that can be used to further study ethnic and regional variations in chronic disease and their association with health care use (18). The Aboriginal Peoples Survey, which was initiated by the federal government in 2001, addresses gaps in other sources, including the lack of data about some determinants of Aboriginal health (eg, housing, employment, residential mobility). More health policies are needed that are culturally sensitive and that focus on the delivery of preventive interventions (23).

The findings of our study, which suggest an overall increasing prevalence of chronic disease in Aboriginal populations and negative shifts in the distribution of risk factors, underscore the importance of community-based primary preventive interventions at the community and primary care level in northern communities.