Dr D Warburton, Cardiovascular Physiology and Rehabilitation Laboratory, University of British Columbia, 6108 Thunderbird Boulevard, Vancouver, Canada V6T 1Z3. E-mail: firstname.lastname@example.org
Overweight and obesity among Aboriginal populations has been a growing challenge within Canada. This investigation aimed to identify the prevalence of overweight and obesity within British Columbian adult Aboriginal populations including both on and off reserve individuals through direct measurement. Further, this study stratified the variations in these rates according to age, gender and geographic region. Weight, height and waist circumference (WC) were measured via standardized procedures, and body mass index (BMI) was calculated. The mean body composition indicators were above recommended for men (BMI = 30.3 ± 5.6 kg m−2; WC = 104.2 ± 14.7 cm) and women (BMI = 30.9 ± 7.2 kg m−2; WC = 99.3 ± 17.1 cm), respectively. Rates of obesity for men and women were similar, 48.4% and 48.7%, respectively, and showed significant increases with age. Abdominal obesity (AO) was significantly greater among female participants, 69.0% compared with 52.7% in men, while both genders also demonstrated significant increases in AO with age. Both obesity and AO rates were found to be significantly greater in the Northern and Interior regions of the province in comparison to the Vancouver/Lower Mainland region. Alarmingly high rates of obesity and AO were observed in this population in men and women at every age and geographic region.
The health of Canadian Aboriginal populations has declined dramatically over the past 60 years (1,2). Traditionally, Canadian Aboriginal populations experienced active lifestyles centered on hunting, gathering, fishing and/or farming societies (3,4). Over the past four to five generations, cultural shifts experienced by this population include decreases in physical activity levels and alterations in diet (2,5). The current epidemic trend of sedentary behaviour occurring within Canada may be two- to threefold higher within the Aboriginal population (6).
Because of these recent changes in diet and lifestyle, Aboriginal populations throughout Canada currently experience epidemic rates of diabetes, obesity and cardiovascular disease (CVD), as high as threefold greater than the Canadian general population (5,7). The rates of these health conditions experienced within this population, mirror trends observed in developing countries, where similar lifestyle changes are occurring (8–10). Past studies of Canadian Aboriginal populations have reported rates of obesity as high as 47% and overweight or obesity prevalence as high as 90% (11,12). Similarly, Aboriginal Canadians have been reported to experience higher rates of abdominal obesity (AO) in comparison to Canadians of European descent (13). It remains to be determined why persons of Aboriginal descent have a greater susceptibility to a variety of chronic conditions.
Previous studies have reported high rates of obesity and overweight status in Aboriginal populations (9,11,12,14,15). However, previous studies using direct measurement of obesity have investigated a specific Aboriginal population in a specific geographic location (14–17). Conversely, studies reporting the obesity of Aboriginal populations on a more global scale have largely utilized self-reported data, and generally exclude on reserve populations (9,11,13,18,19). To date, no studies have conducted direct measurements of obesity including both on and off reserve populations on a more global scale. The primary purpose of this study was to accurately examine the current overweight and obesity prevalence of British Columbian Aboriginal populations. In particular, we sought to determine the prevalence rates of overweight and obesity in Aboriginal adults. In order to identify Aboriginal populations most at risk, and thus with the greatest need for intervention, obesity and AO rates among Aboriginal adults were characterized according to geographic location/residency (urban/rural and on/off reserve), age group, and diabetic and hypertensive status. We hypothesized that the obesity rates observed through direct measurement would be higher than national reports because of the inclusion of on reserve populations, indicating an underestimation of obesity rates by national reports. Additionally, obesity rates were hypothesized to increase among on reserve and rural populations, older populations, and populations carrying health burdens such as diabetes and hypertension.
From January 2007 to February 2010, 759 Aboriginal adults, ≥18 years, from 22 locations around the Canadian province of British Columbia were measured for obesity and AO. Participants represented a range of ages, from 18 to 77 years. At each location, individuals residing both on and off local reserves in nearby towns and cities were tested. Ethics approval was obtained through the Clinical Research Ethics Board at the University of British Columbia and written informed consent was obtained from each participant prior to data collection. All participants were recruited through local First Nations band offices and/or Native Friendship Centres. Testing was conducted within the local Aboriginal communities with the assistance of local Aboriginal healthcare professionals and leaders. As in the past, our research was participatory in nature with Aboriginal community leaders playing important central roles in the design and implementation of the study.
Individual characteristics collected included age, gender and self-report of smoking status and medical personal and/or familiar history of diabetes, CVD and hypertension. Anthropometric measures including height, weight and waist circumference (WC) were assessed (and body mass index [BMI] calculated) according to Canadian Physical Activity, Fitness and Lifestyle Approach standardized protocols (20). To ensure the accuracy and reliability of measurements, CSEP-Certified Exercise Physiologists® were involved in the data collection. Measurements of height were determined using a Seca 214 portable height rod stadiometer (seca corp., Ontario, CA, USA) to the nearest 0.1 cm. Height was measured from heels to the top of the head, with shoes removed while participants were at peak inhalation. Weight was recorded in kilograms to the nearest 0.2 kg using a Health o meter Professional model 320KL scale (Sunbeam Products Inc, Ontario, Canada) while participants wore light clothes having removed any heavy objects, shoes and jackets. Waist circumference was measured at the midpoint between the 12th rib and the iliac crest on the right side of the body using a flexible plastic tape to the nearest 0.1 cm. Participants were instructed to cross their arms over their chest and WC was measured at the end of exhalation. The BMI was calculated as weight in kilograms over height in metres squared. Obesity classifications were determined using the World Health Organization (WHO) principle cut-off points (21). The AO classifications utilized include both WHO level 1 (men ≥ 94 cm, women ≥ 80 cm) and level 2 (men ≥ 102 cm, women ≥ 88 cm) classifications (22). For the purposes of this investigation, WHO level 1 AO ranges were labelled low cut-off, and level 2 ranges were labelled normal cut-off (22). The prevalence of hypertension status was also verified through the direct assessment of resting blood pressure (BP-TRU, VSM Medical, Vancouver, Canada). Geographic locations were classified into four regions. These regions reflect the health authority regions determined by the province of British Columbia, with Vancouver Coastal and Fraser Health authorities combined to form the Vancouver/Lower Mainland region (23).
Statistical analyses were performed using Statistica 9.0 (Stats Soft, Tulsa, OK, USA). For both genders, each age group and all geographic regions, mean, standard deviation and proportion of obesity and AO were determined. Multiple linear regression analysis was conducted to identify correlates of obesity and relative risk ratios were calculated from these regression results (24). For all statistical analyses, P-values < 0.05 were considered significant.
The average age of male and female participants was similar, with a large age span (18 to 77 years) (Table 1). Both genders presented average BMI values in the obese range, with no significant difference between genders. Male participants presented with significantly higher mean WC measures, although both male and female average WC measures fell within the respective abdominal obese category at the normal cut-off range (men ≥ 102 cm, women ≥ 88 cm). Male participants also presented with significantly higher rates of diabetes and directly measured hypertension. It should be highlighted that a significant proportion of the population with high blood pressure (assessed directly), 50.0% of men and 35.2% of women, were unaware of having hypertension.
Table 1. Obesity characteristics and health status of British Columbian Aboriginal participants (mean ± SD)
(n = 182)
(n = 577)
BMI, body mass index; SD, standard deviation; WC, waist circumference.
Nearly half of this population was observed to be obese (≥30.0 kg m−2) with a further third presenting BMI values in the overweight (25.0–29.9 kg m−2 range) according to WHO classifications, as outlined on Table 2(21). Therefore, approximately 78% of the population were either overweight or obese. The prevalence of AO was also high, with two-thirds determined to be centrally obese at the normal cut-off (men ≥ 102 cm, women ≥ 88 cm) (22). Overall, more than 84% of participants sampled were determined to be abdominally obese at the lower AO cut-off range (men ≥ 94 cm, women ≥ 80 cm) (22). Table 2 outlines the prevalence of obese, overweight and abdominally obese status by gender.
Table 2. The prevalence (95% CI) of overweight and obesity among British Columbian Aboriginal populations by gender
Younger individuals presented with lower prevalence of obesity and AO, with a general increase in both obesity and AO with age (Figs 1 and 2). These trends were observed for both male and female participants. Peak obesity and AO prevalence rates were observed among both male and female participants at ≥60 years. Mean BMI and WC values were also observed to increase with age, demonstrating similar peak values in the oldest age group.
In general, relative risks of obesity, overweight or obesity and AO at both cut-off ranges significantly increase with age. Relative risk ratios for obesity and overweight or obesity status significantly increased from age ≥ 40 years in comparison to 18–29 years age groups (Fig. 3). Similarly, relative risk ratios for AO, at both the normal and low cut-off values were found to significantly increase from age ≥ 30 years relative to the risk experienced by 18–29 year individuals (Fig. 4).
Hypertensive individuals demonstrated significantly increased rates of obesity and central obesity after adjusting for age (Tables 3 and 4). Similarly, age-adjusted rates of hypertension significantly increased with increase in obesity status from normal weight for height to overweight to obese. Hypertension rates were also observed to increase significantly in individuals with AO. Obesity rates and AO rates adjusted for age were significantly greater among diabetic individuals compared with non-diabetic individuals. Rates of diabetes prevalence were significantly greater among obese individuals and individuals with AO, after adjusting for age. According to the geographic region, the prevalence of obesity and AO was lowest in the Vancouver/Lower Mainland metropolis region and greatest in the Northern region of the province (Tables 3 and 4). An increase in the prevalence of obesity and AO was observed as geographical regions moved further from the Vancouver/Lower Mainland area, even after adjusting for age and gender differences between regions. Overall, significant increases in age and gender adjusted AO rates were observed from the Vancouver/Lower Mainland to both the Interior region and the Northern region. Similar rates of obesity, overweight or obesity and AO at the low cut-off values were observed between rural and urban residents. Rural residents demonstrated near significantly greater rates of AO at the normal cut-off in relation to urban residents. No significant differences in age and gender adjusted rates of obesity, overweight or obesity, or AO were observed between on and off reserve residences.
Table 3. Correlates of obesity among Canadian Aboriginal adults – multiple logistic regression analysis
Overweight or obese
RR (95% CI)
RR (95% CI)
Adjusted for age.
Adjusted for age and gender; CI, confidence interval, RR, relative risk ratio.
To date, no studies have reported obesity prevalence in Canadian Aboriginal populations based on direct measurement, including both on and off reserve populations and covering a large geographic area including multiple distinct Aboriginal Nations. This study is unique in presenting directly measured obesity rates in a representative sample including both on and off reserve individuals and a large sample of distinct Aboriginal communities. While many investigations utilize self-reported height and weight to evaluate BMI, self-reported height is often overestimated while self-reported weight is often underestimated (25). Overall, these inaccuracies in self-reporting often lead to underestimations of BMI and accordingly underestimated rates of obesity and AO (25). Most investigations reporting obesity rates among the Canadian Aboriginal population are based on Canadian Community Health Surveys or other government initiatives which exclude on reserve populations and often utilize self-reported data (11,18,19). Conversely, local investigations generally focus on one or a few specific Aboriginal populations (1,5,16,17). Recent studies reporting obesity rates among Canadian Aboriginal populations have reported varying rates ranging from 15% to 47.5% depending on the inclusion of on reserve populations and the use of direct measurement (11,19). Without utilizing direct measurement including all sectors of the Aboriginal population, a true report of the obesity prevalence in this population cannot be determined.
This investigation observed equal or greater obesity rates than those previously published, including 48.4% in men and 48.7% in women. While 48.6% of participants were recorded to be obese in this study, a further 33.0% of men and 28.4% of women were overweight, leaving only 18.6% men and 22.9% of women demonstrating a healthy body mass for their height (21). The prevalence of AO in this population is even further alarming with 52.7% of men and 69.0% of women presenting as abdominally obese at the normal cut-off levels. When including the 22.5% of men and 17.8% of women who presented with WC values within the low cut-off range for AO, only 24.8% of men and 13.2% of women experience a healthy WC. The elevated obesity rates observed in this study indicate the underestimation provided by national surveys using self-reported data and excluding on reserve populations.
Obesity rates among Aboriginal populations from this investigation are much greater than rates among the general population. Canadian estimates report the obesity rates in Canada at 14.9%, including 13.9% of women and 15.9% of men (26). Further, 33.3% of Canadians were reported to be overweight, including 41.0% of men and 25.7% of women (26). Similar trends are observed for AO, where reported non-Aboriginal rates are significantly lower than measured rates among Aboriginal populations (1). Reported rates of AO among non-Aboriginal populations are estimated at 26.0%, a much lower rate than the 65% recorded among these Aboriginal Hearts in Training participants (1).
Most national investigations of Aboriginal populations exclude on reserve populations (18,19). Throughout published literature reporting national Aboriginal obesity rates, most investigations have reported lower obesity rates and mean BMI and WC values than those observed in this study. From the January 2000 and June 2005 Canadian Community Health Surveys, an increase in obesity rates of Canadian Aboriginal adults from 22.7% to 25.3% was reported along with overweight increases from 30.1% to 30.9% (18,19). A Canadian Community Health Report published in 2000 reported obesity rates among this population at 29.5% in men and 37.3% in women (18). These studies reporting lower obesity rates, exclude on reserve populations from their sample. Conversely, the 2003 Canadian Community Health Survey, which utilized random telephone dialling, and did not exclude on reserve populations, reported Aboriginal obesity rates most similar to this investigation at 47.5% (11). As reports excluding on reserve populations have reported substantially lower obesity rates than studies including these populations, future investigations reporting on this population need to include both on and off reserve populations to determine a truly representative measure.
While no significant differences in age and gender adjusted rates of obesity and overweight or obese status were observed between on and off reserve residences, on reserve residences presented with greater rates. Obesity rates among off reserve participants included 43.3% with obese BMI measurements and 73.5% with overweight or obese BMI measurements. In comparison, on reserve participants were observed to experience obesity rates including 54.7% obese and 80.0% overweight or obese. A recent Health Canada report identified the obesity rate among First Nations on-reserve populations at 36.0%, with a further 37.0% being overweight (27). Rates of AO are similar to obesity rates. On reserve participants presented with greater rates of AO at both the normal and low cut-off values, although no significant differences in age and gender adjusted relative risks for AO were observed between on and off reserve participants. On reserve participants presented with AO rates of 66.3% and 85.3% at the normal and low cut-off ranges, respectively. Similarly, off reserve participants presented with 57.2% and 80.8% AO at the normal and low cut-off ranges, respectively. These alarmingly high rates of obesity and AO highlight the urgent need for intervention to address the obesity epidemic among both on and off reserve Aboriginal peoples.
Other investigations conducted previously within the province of British Columbia have generally reported lower obesity rates to those observed in this study. The average BMI observations of Interior Aboriginal residents in this study included 30.3 ± 7.1 kg m−2 among women and 30.8 ± 5.4 kg m−2 among men. These values are higher than those observed by Daniel and colleagues roughly 10 years ago in an investigation of similarly aged Interior British Columbian Aboriginal adults (16,17). Daniel and colleagues reported average BMI values of 28.1 ± 5.7 kg m−2 among men and 28.9 ± 5.7 kg m−2 among women (16,17). Another investigation of a younger sample in the British Columbian Interior reported average BMI measurements of 26.99 kg m−2 among Aboriginal participants (15). The lower reports of obesity in previous studies of similar populations indicate the obesity rates are continuing to increase within this population.
Within the Vancouver/Lower Mainland region, average BMI measurements in this investigation of 27.2 ± 4.4 kg m−2 in men and 28.5 ± 6.6 kg m−2 in women were similar to those observed in the recent Multi-cultural Community Health Assessment Trial (M-CHAT) study of individuals from the same region, 28.7 ± 4.3 kg m−2 and 27.7 ± 5.5 kg m−2 for men and women, respectively. The M-CHAT study also observed comparable average WC measurements of 97.2 ± 9.9 cm and 92.1 ± 12.8 cm, for Aboriginal men and women, respectively, similar to 96.2 ± 12.5 cm and 93.5 ± 16.3 cm observed in this study (28,29). Similar findings of two independent studies indicate these obesity values are likely accurate.
Similar rates of obesity and overweight status were observed among Aboriginal residents of rural and urban locations. However, while significant differences in AO were not observed, rural residents (75.2% at normal cut-off range) demonstrated higher rates of AO in comparison to their urban counterparts (56.0% at normal cut-off range). We observed 56.0% obesity among rural residents, with a further 28.4% overweight. These results match with rural and remote Aboriginal communities from the neighbouring province of Alberta, where obesity rates of 55.0% and a further 29.4% overweight were found (30). Alberta rural and remote locations also demonstrated 76.8% AO, in line with the 76.0% AO in rural British Columbian communities (30). Similar findings in two separate provinces indicate the obesity epidemic experienced among Aboriginal British Columbians may be similarly experienced by neighbouring provinces. These alarmingly high rates of obesity and AO observed in rural and remote locations support the urgent need for intervention and further health initiatives within these communities, throughout the country.
This study observed no significant difference in obesity or overweight prevalence between men and women, highlighting the need for interventions in both genders. The similar rates among men and women are comparable to previous national reports (31). However, as previously reported, women were observed to experience significantly greater prevalence of AO in comparison to men, even after adjusting for age (5,32). In contrast to the January 2000 and June 2005 Canadian Community Health Surveys, this British Columbian study observed a significant increase in the prevalence of obesity with age (19). These increases in obesity with age have previously been reported among Canadian Aboriginal populations in the 2004 Canadian Community Health Survey (31).
Obesity is known to be associated with hypertension, throughout the lifespan (33,34). Results of this study support this observance, where individuals with hypertension were observed to have significantly increased rates of obesity as well as AO, after adjusting for age and gender. Consistent with Kotsis et al. (33), obese individuals experienced significantly increased rates of hypertension in comparison to individuals of normal weight for height.
Within this investigation, diabetics were observed to have significantly higher obesity rates, after adjusting for age and gender. Since the 1970s, links between obesity and diabetes have been reported (35). Obesity and weight gain have been associated with increased prevalence of type 2 diabetes (36). The high prevalence of obesity within this population indicates a substantial portion of this population is at risk for future development of diabetes.
Many trends in obesity distribution previously published were similarly observed in this investigation. Trends for geographic region in this study are similar to those observed for the general population, with the lowest obesity rates in Vancouver, although the 34.5% obesity rates for Aboriginal populations in the Vancouver region are substantially greater than the 6.2% reported for the general population (11). This investigation observed significantly increased rates of obesity and AO in both genders with geographic distance from the Vancouver/Lower Mainland region, even after adjusting for age. Other studies have also observed greater obesity rates for Northern regions of the country, usually defined as the Territories (26). This trend of increasing obesity with more Northern locations appears to exist within the province as well, with the Northern region demonstrating significantly increased obesity and AO rates from all other regions in the province.
Despite rare obesity instances among ancestral First Nations communities, obesity rates among current Aboriginal populations are alarmingly high (11). Increasing rates of obesity among this population have been linked to increased physical inactivity and dietary changes (17). Among Aboriginal populations, BMI can be used as a standard predictor of diabetic status, A1C concentrations and plasma glucose concentrations, and AO is associated with many CVD conditions (17,37). The high rates of obesity and AO measured in this investigation highlight an alarming risk for diabetes and CVD among this population. The associated alarming risk for diabetes and CVD among this population indicates the urgent need for interventions addressing the obesity and AO burden among this population to prevent further development of diabetes and CVD within Aboriginal adult populations.
The present data indicates alarmingly high rates of obesity and AO across genders, all ages and all geographic regions in adult Aboriginal populations. Because of direct measurement with the use of CSEP-Certified Exercise Physiologists® and Canadian Physical Activity, Fitness and Lifestyle Approach standardized protocols, these results are more accurate than many other investigations based on self-reported data (20). Previously published studies have generally reported lower obesity and AO rates among this high-risk population. The discrepancy between these results and previous studies underscores the need for direct measurement, including on reserve populations in future estimates of obesity rates among Aboriginal populations. The high rates of obesity presented here indicate the obesity epidemic among Canadian Aboriginal populations continues. These results highlight the need for further interventions to address obesity and other chronic conditions experienced by this population. Increasing rates of obesity and AO with age, particularly in women, highlight the need for interventions across the lifespan. Interventions to address this obesity epidemic should address the changes in diet and lifestyle recently experienced within this population.
Studies including only Aboriginal populations from metropolitan or off reserve locations underestimate obesity rates among this population. Reports of obesity within this population need to utilize direct measurement and include all geographic, on and off reserve locations to accurately determine the extent of the obesity epidemic. Obesity remains a concern among this high-risk population and further initiatives are required to combat this epidemic.
Conflict of Interest Statement
We have no conflict of interest to declare.
This research was supported by SportMed BC, the Healthy Heart Society of BC, ActNow BC, the Physical Activity Support Line, the Canada Foundation for Innovation, the BC Knowledge Development Fund, the Canadian Institutes of Health Research (CIHR), the Michael Smith Foundation for Health Research (MSFHR) and the Natural Sciences and Engineering Research Council of Canada. Dr Warburton was supported by a CIHR New Investigator Award and a MSFHR Clinical Scholar Award. Heather Foulds was supported by grants from the National Aboriginal Achievement Foundation, the Foundation for the Advancement of Aboriginal Youth, a University of British Columbia Special Graduate Scholarship, a Dofasco Inc First Nations Fellowship and a Pacific Century Graduate Scholarship.