The prevalence of obesity in the study population is among the highest reported for a Canadian First Nations community on a reserve (6
) and is substantially higher than that among the general Canadian (4
) and off-reserve Aboriginal populations (4
), whether the studies used self-reported data (4
) or direct measurement (15
). The high prevalence of obesity in the study population is concerning given the etiologic role of obesity in diabetes, heart disease, stroke, and some cancers. The prevalence of diabetes that we found is one of the highest reported among Canadian First Nation populations (6
One finding of concern is the high prevalence of obesity among young adults, especially young women of reproductive age. The relationships between maternal obesity and gestational diabetes, type 2 diabetes, poor birth outcomes, and development of obesity and type 2 diabetes among offspring are well documented (17
). Thus, the prevalence of obesity in this young study population warrants intervention. These findings are important for 2 reasons: 1) participants developed chronic conditions at young ages, and 2) hypertension and diabetes cases were undiagnosed among a large proportion of obese participants.
Results from logistic regression confirmed established associations between obesity and plasma lipid levels, hypertension, insulin resistance, and sociodemographic factors in the study population. The sex-specific regression analyses did not include lipids for abdominal obesity among women. We offer 2 possible reasons for this. First, the prevalence of abdominal obesity was high among women in all age groups but the presence of abnormal lipid levels was not. These age differences may have been blunted because our outcome (obesity) was present in all age groups. Second, previous research has shown significant sex differences in the relationship between adiposity and plasma lipids (21
). Because abnormal lipid levels did occur among women, this finding warrants further examination.
We found a high prevalence of comorbidity even among the youngest age groups. The Diabetes and Related conditions in Urban Indigenous people in the Darwin region (DRUID) study also found high numbers of cardiovascular comorbidities among Australian Aborigines, and a higher number of comorbidities with increasing age (22
). A large proportion of the study participants had undiagnosed diabetes and hypertension, despite the known strong correlations among obesity, diabetes, dyslipidemia, and hypertension (23
) (we could not determine the extent of undiagnosed dyslipidemia among study participants because we did not ask them to self-report abnormal lipid levels). In a previous study, risk factors for not having blood pressure measured included male sex, never being married, not having a regular physician, being younger, and belonging to an Aboriginal or other ethnic minority group (24
). In our study, the likelihood of not having hypertension diagnosed was higher for men (OR, 3.27; 95% CI, 1.74-6.10; P
< .01) and younger participants (OR, 1.04; 95% CI, 1.01-1.07; P
In our study, the undiagnosed hypertension was not benign. The extent of comorbidity among participants with newly diagnosed hypertension was similar to that for those with previously diagnosed hypertension. In addition, the risk for microalbuminuria was significantly higher among participants with newly diagnosed hypertension compared with those without hypertension but not significantly different between those with newly diagnosed hypertension and those with previously diagnosed hypertension. This suggests that newly diagnosed hypertension among participants had existed for some time. The association between hypertension and outcomes such as CVD and stroke warrants vigilant screening on the part of health care providers, especially in high-risk populations. Some participants in our "newly diagnosed" group may have been told by a physician that they did have hypertension, but they may not have remembered or they may have not understood. However, none were receiving antihypertensive treatment, so they probably had not received a hypertension diagnosis before our study.
The study is subject to limitations. First, our sample was based on volunteers and therefore may not be representative of the community as a whole or of other Canadian First Nations communities. A screening study based on a volunteer sample may attract primarily healthy people who are motivated to learn more about their health, resulting in an underestimation of illness. On the other hand, a screening study can attract people who already have health problems and are seeking additional medical assistance, which may result in an overestimation of the prevalence of illness in a population. We do not think our sample was overrepresented by either group because men and women were equally represented, and the age distribution of our sample matched that of the eligible population (10
). Another indication that the prevalence of illness in the community was not overstated is that only half of the community members known to have diabetes participated in the study. None of the 15 people with end-stage renal disease participated, and only 3 of 10 community members with amputations participated (10
). The prevalence of chronic disease and risk factors that we report are not substantially out of line with previous research.
A second limitation is the use of a fasting glucose test rather than a glucose tolerance test. More people with diabetes may have been identified if 2-h glucose tolerance tests were conducted. However, our protocol is acceptable for epidemiologic research. A third limitation is that we did not validate the self-reported hypertension or diabetes status measures with local health care providers, so we may have underestimated self-reported prevalence and therefore overestimated undiagnosed cases. However, we have previously reported lack of adherence with standards of care in this community in relation to foot examinations among people with diabetes (10
), so participants may not have been tested for diabetes and hypertension even when indicated. Finally, the study is cross-sectional, so we cannot infer the temporal sequence of events.
The prevalence of obesity in this population is among the highest reported among Canadian First Nation populations, particularly among women in their reproductive years. The extent of obesity-related comorbidity in this population is high even among young adults, and women at almost every age have a significantly higher rate of comorbidity than do men. A sizable proportion of participants have undiagnosed hypertension that may have been present for some time, given the significant associations with the other chronic diseases and microalbuminuria. The prevalence of cardiovascular and renal disease risk factors in this population may portend a larger prevalence of cardiovascular and renal disease. In addition, given the influence of maternal obesity and diabetes on the health of offspring, an increase in childhood obesity and type 2 diabetes could occur in the community.
An increasing prevalence of obesity and obesity-related conditions is not inevitable, however. Many prevention activities are under way. First, a research intervention in the community is focused on preventing gestational diabetes through controlling weight gain during pregnancy with exercise and diet. Second, the community operates a fitness center that has good equipment and instruction. Third, the health center offers education on diet, exercise, and wellness. Fourth, walking groups for youth and adults are organized through the health center. Fifth, activity programs for young people operate out of the local schools. However, given the well-established effect of obesity on health, continued surveillance of chronic disease and risk factors is warranted, as are further health promotion and health education initiatives. We continue to work with the community to develop and evaluate primary and secondary prevention activities.