PREVALENCE OF DIABETES MELLITUS IN THE ABORIGINAL AND
NON-ABORIGINAL PEOPLE LIVING IN THE BELLA COOLA VALLEY
by
Jana Patenaude
BSc., University of Northern British Columbia, 2001

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
in
COMMUNITY HEALTH SCIENCE

THE UNIVERSITY OF NORTHERN BRITISH COLUMBIA
March 2004

Jana Patenaude, 2004

1^1

Library and
Archives Canada

Bibliothèque et
Archives Canada

Published Heritage
Branch

Direction du
Patrimoine de l'édition

395 W ellington Street
Ottawa ON K 1A 0N 4
Canada

395, rue W ellington
Ottawa ON K 1A 0N 4
Canada
Your file Votre référence
ISBN: 0-494-04678-3
Our file Notre référence
ISBN: 0-494-04678-3

NOTICE:
The author has granted a non­
exclusive license allowing Library
and Archives Canada to reproduce,
publish, archive, preserve, conserve,
communicate to the public by
telecommunication or on the Internet,
loan, distribute and sell theses
worldwide, for commercial or non­
commercial purposes, in microform,
paper, electronic and/or any other
formats.

AVIS:
L'auteur a accordé une licence non exclusive
permettant à la Bibliothèque et Archives
Canada de reproduire, publier, archiver,
sauvegarder, conserver, transmettre au public
par télécommunication ou par l'Internet, prêter,
distribuer et vendre des thèses partout dans
le monde, à des fins commerciales ou autres,
sur support microforme, papier, électronique
et/ou autres formats.

The author retains copyright
ownership and moral rights in
this thesis. Neither the thesis
nor substantial extracts from it
may be printed or otherwise
reproduced without the author's
permission.

L'auteur conserve la propriété du droit d'auteur
et des droits moraux qui protège cette thèse.
Ni la thèse ni des extraits substantiels de
celle-ci ne doivent être imprimés ou autrement
reproduits sans son autorisation.

In compliance with the Canadian
Privacy Act some supporting
forms may have been removed
from this thesis.

Conformément à la loi canadienne
sur la protection de la vie privée,
quelques formulaires secondaires
ont été enlevés de cette thèse.

While these forms may be included
in the document page count,
their removal does not represent
any loss of content from the
thesis.

Bien que ces formulaires
aient inclus dans la pagination,
il n'y aura aucun contenu manquant.

Canada

Abstract

The purpose of this study was to determine the prevalence rate of diabetes among the
aboriginal and non-aboriginal populations of the Bella Coola Valley, British Columbia, as
well as to explore the relationship between age, gender, weight, body mass index and
aboriginal status, and the risk of developing type 2 diabetes in this community. A
retrospective chart review was conducted on all residents living in the Bella Coola Valley as
of September 2001 who had a clinic chart at the Bella Coola Medical Clinic. The ageadjusted prevalence of diabetes among the aboriginal population was 12.5%. Among the
non-aboriginal population, the prevalence rate was similar to what has been reported for the
general population of Canada (4.8%). It was determined that age, weight, body mass index
and aboriginal status were all significant contributors to the risk of developing type 2
diabetes. The results of this study indicate that diabetes is a significant health issue that
needs to be addressed in the Bella Coola Valley.

11

Table of Contents

Page
Abstract

ii

Table of Contents

iii

List of Tables

iv

List of Figures

V

Acknowledgements

vi

Chapter One

Introduction

1

Chapter Two

Literature Review
Diabetes Mellitus
Diabetes and Aboriginal Populations
Theories Behind the High Prevalence of
Type 2 Diabetes in Aboriginal Populations
Research Questions

4
4
10
16

Methods
Research Design
Study Community
Bella Coola Valley Population
Health Services in the Bella Coola Valley
Data Collection - The Bella Coola Valley
Clinic Population
Ethics Approval
Data Analysis

22
22
22
26

Chapter Three

21

29
29
32
33

Chapter Four

Results

35

Chapter Five

Discussion

45

Chapter Six

Implications of this Study

56

Chapter Seven

Future Research

58

References

59

Appendix I

Ethics Approval Certificate

65

Appendix II

Direct Age-adjustment of Prevalence Rates

67

Ill

List of Tables

Table 1.

Diabetes Prevalence Rates for Aboriginal Populations
in Canada, West to East

12

Table 2.

Age and Gender Breakdown of the Bella Coola Valley
Based on the 2001 Census (Statistics Canada, 2001)

26

Tahle 3.

Number of People Living On and Off Reserve in the
Bella Coola Valley, 2001 Census (Statistics Canada, 2001)

28

Tahle 4.

Age and Gender Breakdown of the Bella Coola Valley
Study Population

31

Table 5.

Comparison of 2001 Census Population and Study Population
Broken Down by Age and Gender, for the Bella Coola Valley

32

Table 6.

Summary of the Characteristics of the Adult Study Population

35

Table 7.

Odds Ratios (OR), p Values and 95% Confidence Intervals (Cl)
for Diabetes from Univariate and Multivariate Logistic
Regression Models

42

IV

List of Figures

Figure 1.

Map of British Columbia highlighting the locations of
Bella Coola, Williams Lake and Vancouver.

24

Figure 2.

Detailed map of the Bella Coola Valley.

25

Figure 3.

Map of the Central Coast Regional District Electoral Areas.
Areas C, D, and E were used to calculate the census population
for the Bella Coola Valley.

27

Figure 4.

The prevalence of type 2 diabetes in aboriginal and
non-aboriginal males based on age.

37

Figure 5.

The prevalence of type 2 diabetes in aboriginal and
non-aboriginal females based on age.

37

Figure 6.

Prevalence rates of type 2 diabetes among aboriginal and
non-aboriginal adults based on weight (kg).

39

Figure 7.

Prevalence rates o f type 2 diabetes among aboriginal and
non-aboriginal adults based on body mass index (kg/m^).

40

Acknowledgements

I wish to acknowledge several people for their support and guidance in completing
this thesis. Dr. Harvey Thommasen for the time and care he put into developing the dataset
for this thesis, and the continued support he provided me throughout the process of
developing my thesis; Don Voaklander for his expertise in epidemiological studies; Sylvia
Barton for her attention to detail; and Amy MacArthur, Research Coordinator for the UBC
Department of Family Practice, for her help with my understanding of the statistical analysis
of the data.
I would like to acknowledge and thank the people of the Bella Coola Valley for
allowing this research to be conducted and for taking an active role in improving the health
of their community.
I wish to acknowledge the support I have received through the Western Regional
Training Centre for Health Services Research. In addition to financial support, the
opportunities that they have provided through workshops, networking and a field placement
have given me experiences and skills that eould not be gained elsewhere.
Finally, I would like to acknowledge the support of my family and friends. To my
parents, who instilled in me the desire to learn and provided me with unconditional love and
support. To Jared, who loved me enough to force me to stop procrastinating and actually get
on with the business of writing this thesis, and to my fiiends, who were always there for me
whether it was to bounce ideas off them or just to escape the fiustrations that inevitably come
with writing a Master’s thesis.

VI

Chapter One
Introduction

Type 2 diabetes has gone from being nearly nonexistent in the North American native
population in 1940 to almost epidemic proportions 50 years later (Evers, McCracken,
Antone, & Deagle, 1987; Pioro, Dyck, & Gillis, 1996). As the traditional lifestyle of many
aboriginal groups is being abandoned in favour of a more “modem” lifestyle, the incidence
of chronic disease is increasing and is becoming one of the major health concerns for
aboriginal people. The change in diet and decrease in activity that comes along with the
“modem” lifestyle are two of the major factors many researchers feel contribute to the rapid
increase in the prevalence of type 2 diabetes in aboriginal populations (Brassard, Robinson,
& Lavallee, 1993). Type 2 diabetes is of particular eoneem to aboriginal people because of
the earlier age of onset, the greater severity at diagnosis and the higher rate of complications
seen in this population (Health Canada, 2000).
In the general population, if diabetes is diagnosed in children and/or adolescents it is
primarily type 1 diabetes, an autoimmune disorder in which the pancreas produces little or no
insulin (Health Canada, 2002). It has been determined that within aboriginal populations
type 1 diabetes is very rare, however, the diagnosis of type 2 diabetes among children and
adolescents is becoming increasingly more common (First Nations and Inuit Health Branch,
2003; Young, 1994). The resulting longer duration of diabetes in these people dramatically
increases the likelihood that they will develop complications earlier in life (Young et ah,
2002). This not only places an increased burden on the individual, but also results in more
strain on an already taxed health care system. The earlier age at onset also means that there
are a greater proportion of aboriginal women of childbearing age who are diabetic. This is of

concern given the link between maternal diabetes and the risk of subsequent development of
type 2 diabetes in the offspring, starting a vicious cycle of type 2 diabetes (Dabelea,
Knowler, & Pettitt, 2000).
The greater severity at diagnosis seen among aboriginal diabetics is often the result of
late detection or diagnosis of diabetes. It bas been reported that aboriginal people seek
medical attention less frequently than the general Canadian population. For example, 67% of
North American Indians saw a General Practitioner at least once in the previous year
compared to the general population average of 82%, according to the 1991 Aboriginal
Peoples Survey (Health Canada, 2000). This lack of contact with medical professionals
combined with the fact that the symptoms of diabetes can be very subtle and easily
overlooked has led to the potential for greater severity of diabetes at diagnosis. Aboriginal
people also experience high rates of complications associated with diabetes (Health Canada,
2002). While all people with diabetes are at risk for developing complications, studies have
found that aboriginal people are more likely to develop complications than the general
diabetic population. For example, in Manitoba it was found that an aboriginal person was
twelve times more likely to develop diabetic nephropathy than a non-aboriginal person
(Health Canada, 2000). Similarly, aboriginal diabetics experience a higher rate of mortality
from diabetes than do the non-aboriginal diabetics (Young, 1994).
According to a report published by Health Canada (2000) the prevalence rate of
diabetes among the First Nations people in Canada is 6.4%, more than double the rate for the
general population of Canada (3.1%). A higher rate was seen in First Nations people living
on-reserve as opposed to those living off (8.5% vs. 5.3%). All of these rates are based on the
responses to the 1991 Aboriginal Peoples Survey. While the prevalence rate of diabetes is

higher in the aboriginal population overall, it is important to note that there is considerable
regional variation. A study conducted by Young, Szathmary, Evers and Wheatley (1990)
showed that the prevalence rates of diabetes varied greatly among the provinces and
territories, with the Atlantic provinces and Ontario having the highest rates, at 8.7% and
7.6%, respectively, while British Columbia, the Yukon and Northwest Territories had the
lowest, at 1.6%, 1.2% and 0.8% respectively. This observed trend in the prevalence rates
indicates a west-east gradient, with the prevalence rate of diabetes increasing as one moves
from western to eastern provinces. This same study, in addition to others, has also pointed
towards a north-south gradient, with the prevalence of diabetes increasing towards the south
(Brassard et al., 1993; Delisle & Ekoe, 1993). One possible explanation for this trend is that
the southern parts of most provinces are generally more urbanized and therefore, it is more
likely that the aboriginal people living in these areas would have adopted a more “modem
lifestyle” (Health Canada, 2000).
Given the high prevalence of diabetes seen in the eastern provinces’ aboriginal
populations, it has been these provinces that have been studied most extensively. By
comparison the prevalence rate of diabetes in the various aboriginal groups in British
Columbia has not been extensively studied. For this reason, the focus of the present study is
to determine the prevalence rate of diabetes among a coastal aboriginal population in British
Columbia, and to add this knowledge to the body of literature of the prevalence of diabetes
among aboriginal populations across Canada.

Chapter Two
Literature Review

To fully understand the prevalenee of diabetes and the impact that prevalence has on
a population, the full extent of the disease, and the associated burden of illness, needs to be
explored. It is only within this context that the importance of measuring the prevalence of
diabetes in a population becomes apparent. The first step is to generally understand diabetes,
the different types, the risk factors, and the complications. Secondly, this knowledge can be
explored within the context of aboriginal populations, noting similarities and differences.

Diabetes Mellitus

Diabetes mellitus is a chronic disease that results from the body’s inability to produce
and/or use insulin properly. Insulin is a hormone that is required by the body to convert
glucose into energy (Health Canada, 2002). Diabetes mellitus commonly manifests itself in
three different forms: type 1, type 2, and gestational. Type 1 and type 2 are both chronic
conditions, though they differ in their pathogenesis, whereas gestational diabetes is a
temporary condition that occurs during pregnancy for some women. In addition to these
types, diabetes can also develop as a result of genetic defects in insulin action or pancreatic P
cell function. Patients bom with these sorts of genetic defects typically present early in life
(e.g. before age 25 years), are non-ketotic, and respond to oral hypoglycemic medications.
These conditions have been called mature-onset diabetes of the young. Diabetes can also
develop as a secondary condition to diseases or disorders of the pancreas such as pancreatitis,
cancer, and other metabolic disorders, trauma or infection (Meltzer et ah, 1998).

Less than ten percent of all people with diabetes have type 1 diabetes, which occurs
when the body is unable to produce insulin. Over ninety percent of people with diabetes
have type 2 diabetes, which occurs when the body cannot produce sufficient insulin and/or is
unable to effectively use the insulin produced. Unlike type 1 diabetes, which is dependent on
daily injections of insulin, type 2 diabetes can often be controlled through diet and exercise,
sometimes with the aid oral medications and/or insulin (Canadian Diabetes Association,
2000). Type 2 diabetes is the focus of this study, and therefore much of the subsequent
information will pertain primarily to type 2 diabetes.
In addition to the three types of diabetes, there are two other conditions, often referred
to as pre-diabetic conditions, which are important to consider. Impaired fasting glucose is a
condition where the fasting blood glucose level is higher than normal, but does not reach the
threshold required for a diagnosis of diabetes (American Diabetes Association, 2003).
According to Canadian clinical practice guidelines for diabetes (Meltzer et al., 1998) a
fasting blood glucose level of between 6.1 mmol/L and 6.9 mmol/L equates impaired fasting
glucose. Impaired glucose tolerance is a similar condition, however it results from higher
than normal blood glucose levels, 7.8 mmol/L to 11.0 mmol/L, 2 hours after a 75 g glucose
load (Meltzer et al., 1998). Both of these conditions place individuals at a higher risk for
developing type 2 diabetes, as well as cardiovascular disease (American Diabetes
Association, 2002; Meltzer et al., 1998).
Prevalence
According to a report issued by Health Canada (2002), 4.8% of Canadians, 20 years
of age and older, had diabetes mellitus in 1998/99. The same report showed that the
prevalence of diabetes increases with age and that men generally had a higher prevalence of

diabetes than women. In British Columbia, the prevalence rate is approximately 4.9%,
almost equivalent to the rate for Canada as a whole (Diabetes Working Group, 2002). There
is considerable agreement among those who study diabetes that the prevalence of the disease
is on the rise, and if left unchecked will continue to rise. In the United States the ageadjusted prevalenee rate o f diabetes has almost doubled in 20 years, from 2.77% in 1980 to
4.48% in 2000 (National Center for Chronic Disease Prevention and Health Promotion,
2003). In British Columbia, it is estimated that the prevalenee of diabetes will reach 7.1% by
the year 2010 (Diabetes Working Group, 2002).
There have been several reasons postulated as to why there is such a marked increase
in the prevalence o f diabetes. The primary reasons include the fact that the population as a
whole is getting older, and the prevalence of diabetes generally increasing with age (Health
Canada, 2002). Another key reason for the dramatic increase in the prevalence of diabetes is
the parallel increase in the prevalence of obesity, a known risk factor for type 2 diabetes
(Choi & Shi, 2001; Diabetes Working Group, 2002; Sheard, 2003). There is one caveat to
keep in mind when considering the increase in the prevalence of diabetes. The diagnostic
criteria for diabetes were changed in the late 1990's when the fasting plasma glucose value
for diagnosis was decreased to 7.0 mmol/L from 7.8 mmol/L. Therefore, more people with
diabetes are being identified (Expert Committee on the Diagnosis and Classification of
Diabetes Mellitus, 2002; Meltzer et ah, 1998). This change should improve the accuracy of
the prevalence rates by decreasing the number of diabetics who go undiagnosed. It is
estimated that in Canada as many as one third of all people with diabetes are undiagnosed
(Health Canada, 2002).

Risk Factors
There has been substantial research done on the risk factors associated with type 2
diabetes. While many risk factors have been examined, there is consensus among the
research community that obesity, physical inactivity, age, race and a family history of
diabetes, constitute the major risk factors. It is important to note that none of these risk
factors work in isolation, but rather it is the interrelationship of these factors that increases
the risk of developing diabetes. The more risk factors that an individual possesses, the
greater the risk for developing diabetes (Fletcher, Gulanick, & Lamendola, 2002).
Obesity is the most recognized risk factor for developing type 2 diabetes. Almost
every study that investigates the risk factors for type 2 diabetes, identifies obesity, especially
abdominal adiposity, or central obesity, as one of the major risk factors (Fletcher et ah,
2002). When reviewing the literature on obesity, and the associated risk for the development
of type 2 diabetes, it is important to note the definition of overweight and/or obese used.
Different studies use varying body mass index (BMI) values to represent either overweight or
obese individuals. For example, the World Health Organization defines overweight as
having a BMI of over 25 kg/m^ and obese as having a BMI of over 30 kg/m^ (Puska, Nishida,
& Porter, 2003). Another study uses a BMI of greater than 27 kg/m^ to signify being
overweight and a BMI of between 25 kg/m^ and 27 kg/m^ as having some excessive weight
(Choi & Shi, 2001). These distinctions become quite important when comparing the results
of different studies.
In a report by Health Canada (2002), it was determined that diabetes is more common
among overweight (BMI > 25 kg/m^) individuals and that the tendency to be overweight
increases with age. Another study, using data from the National Population Health Survey in

Canada, found that the prevalence of diabetes increases with increasing body mass index.
The calculated odds ratio for overweight (BMI > 27 kg/m^) individuals was 3.86, while for
individuals with only some excessive weight (BMI 25 - 27 kg/m^) the odds ratio was 1.60
(Choi & Shi, 2001).
Another risk factor closely associated with obesity, and in turn the development of
type 2 diabetes, is physical inactivity. With the shift to an increasingly sedentary lifestyle the
proportion of individuals who are participating in a less than optimal level of physical
activity is increasing. It has been noted that more women than men reported physical
inactivity in Canada in 1998/99 and that physical inactivity appeared to increase with age.
Approximately 65% of individuals with self-reported diabetes also reported being physically
inactive (Health Canada, 2002). In a study by Choi and Shi (2001) investigating the risk
factors associated with diabetes, the authors looked specifically at the correlation between
energy expenditure (i.e. the amount of physical activity) and the prevalence of diabetes.
They discovered that the prevalence of diabetes increased as the energy expenditure
decreased. For example, inactive males had a diabetes prevalence rate of 4.0%, while active
males had a prevalence rate of only 2.6%. A similar result was seen with females.
When examining the prevalence rates of diabetes, a clear trend of the rate increasing
with age can be seen in almost any population (Choi & Shi, 2001; Health Canada, 2002;
Zimmet, 1982). Type 2 diabetes was previously known as ‘adult-onset’ diabetes, given that
individuals generally developed the disease later in life. In fact the age of 45 years was used
as an important cut-off when looking at the prevalence of type 2 diabetes, however, in recent
years the prevalence of this type of diabetes has been rapidly increasing in the younger age
groups (e.g. 30 - 39 years) (Fletcher et ah, 2002).

Various studies have shown that different ethnie groups have varying prevalences of
diabetes within their populations. For example, minorities in the United States experience a
diabetes prevalence rate that is two to six times higher than that for Caucasians (Fletcher et
al., 2002). It has been proposed that certain ethnic groups have a genetic predisposition to
diabetes that is uncovered by exposure to partieular environmental factors (Arslanian, 2002).
This theory will be examined further within the context of diabetes and aboriginal
populations.
Supplementary to the role genetics play in the predisposition of certain ethnic groups
to diabetes, genetics also predisposes some individuals to diabetes if they have a family
history of the disease. It has been determined that type 2 diabetes is an inherited condition
that is expressed under certain environmental factors. People who have a single parent with
diabetes have a 3.5 fold greater risk of developing diabetes than those who have no parents
with the disease; with two diabetic parents the risk increases to 6 fold (Fletcher et al., 2002).
In a study of type 2 diabetes in children, it was discovered that of the Mexican-American
children with diabetes, 80% had a first degree relative with the disease. For Black children
with diabetes, 95% had a family history of diabetes (Arslanian, 2002).
Complications
Many of the risk factors associated with developing diabetes are also associated with
the development of a variety of complications in people with diabetes. These complications
are often related to high blood glucose levels and an extended duration of diabetes. They can
affect all parts of the body and range from microvascular to macrovascular in nature.
Microvascular complications arise from damage to small blood vessels that result in a
decrease in blood circulation. Some examples of microvascular complications include

retinopathy, nephropathy and periodontal disease. Maerovascular complications are similar
to microvascular, except that it is the larger blood vessels that are damaged. This results in
outcomes such as cardiovascular disease, cerebrovascular disease, stroke, ischemic heart
disease and lower limb amputation (Health Canada, 2002). These complications can have a
devastating effect on the person’s life, affecting both its quality and longevity.
Given the overwhelming prevalence of diabetes, and the potentially devastating
consequences of the disease, appropriate management is crucial. It is generally agreed upon
that appropriate management of diabetes can prevent, or at the very least delay the onset of,
many of the complications associated with the disease (Health Canada, 2002). In addition to
maintaining a healthy lifestyle, including diet and exercise, maintaining optimal control over
blood glucose levels, blood pressure, and blood lipid composition has proven to be effective
in staving off complications and improving the quality of life for someone with diabetes
(Vaaler, 2000).

Diabetes and Aboriginal Populations

There is considerable consensus among the literature that prior to 1940, type 2
diabetes was nearly non-existent in aboriginal people (Evers et al., 1987; Hernandez, Antone,
& Cornelius, 1999; Macaulay, Montour, & Adelson, 1988; Young, McIntyre, Dooley, &
Rodriguez, 1985). Since the end of the Second World War, there has been an explosion in
the number of cases o f diabetes in aboriginal populations to the point where it is considered
an epidemic in many tribes. The prevalence rates of type 2 diabetes vary with different
aboriginal groups and their geographic location, but overall the prevalence rates for
aboriginal populations are generally higher than non-aboriginal populations.

10

Prevalence
The prevalence of diabetes among aboriginal people bas been extensively studied in
the United States, particularly the Pima Indians in Arizona where diabetes is prevalent in
almost 50% of the adult population (Benysbek, Martin, & Johnston, 2001). A study
conducted using the Indian Health Services national outpatient database estimated the ageadjusted prevalence of diabetes among the Native Amerieans and Alaska Natives in 1997 to
be 8.0% (Burrows & Geiss, 2000). An analysis of the prevalence data broken down by age
and sex showed that the prevalence rate of diabetes inereases with age and is higher among
women than men. However, from the years 1990 to 1997, men experienced a greater
increase in their prevalence of diabetes than women, despite the fact that women have a
higher overall prevalenee. One disturbing observation in this study was the fact that for
people under the age of 45 years, the increase in the prevalence rate of diabetes was ten times
greater for Native Americans than for the general United States population. As type 2
diabetes has traditionally primarily affected older people, this increasing trend of an earlier
age of diagnosis is cause for concern. The increased duration of diabetes for these individuals
puts them at a greater risk of developing complications.
While the Native Americans in the United States have been the subject of much
research over the last 40 years or so, studies of the aboriginal populations in Canada were not
done extensively until around the 1980’s. Since then a number o f studies have been
conducted, especially of the aboriginal groups in Ontario and Quebec. Fewer studies have
been done on the aboriginal groups in British Columbia, though more are emerging. Table 1
provides a summary of the various prevalence rates reported for different aboriginal groups
across Canada.

11

Table 1, Diabetes Prevalence Rates for Aboriginal Populations in Canada, W est to East.
Location

Aboriginal
Group

Age
Group

British
Columbia

All ages

British
Columbia

On-reserve
First
Nations
On-reserve
First
Nations
All First
Nations

British
Columbia

All First
Nations

35 years 4-

British
Columbia

On-reserve
Aboriginals

All ages

Alberta

On-reserve
Aboriginals

All ages

Saskatchewan

All First
Nations
groups in
the province
On-reserve
Aboriginals

20 years +

British
Columbia

Saskatchewan

35 years

All ages

All ages

Manitoba

On-reserve
Aboriginals

All ages

Sandy Lake,
Ontario

Ojibwa,
Cree

10 years +

Southwestern
Ontario

Oneida,
Chippewa,
Muncey
Cree

5 years +

Moose
Factory,
Ontario

15 years +

Type of
Prevalence
(year)
Crude
prevalenee
(1997)
Crude
prevalence
(1997)
Crude
prevalence
(1995)
Crude
prevalence
(1995)
Direct agestandardization
(1987)
Direct agestandardization
(1987)
Direct agestandardization
(1990)

Prevalence
Rate

Reference

2.6%

(Johnson,
Martin, &
Sarin, 2002)
(Johnson et
al., 2002)

Direct agestandardization
(1987)
Direct agestandardization
(1987)
Direct agestandardization
(1991)
Direct agestandardization
(1985)
Direct agestandardization
(1998)

3.9%

7.2%

2.2%

6.3%

1.6%

5.1%

9.7%

5.7%

26.1%

(Martin &
Yidegiligne,
1998)
(Martin &
Yidegiligne,
1998)
(Young,
Szathmary et
al., 1990)
(Young,
Szathmary et
a l, 1990)
(Pioro et al..
1996)

(Young,
Szathmary et
al., 1990)
(Young,
Szathmary et
al., 1990)
(Harris et al..
2003)

14.7%

(Evers et al..
1987)

13.1%

(Maberley,
King, &
Cruess,
2000)

12

Table 1. Continued
Location

Aboriginal
Group

Age
Group

Moose
Factory,
Ontario
Ontario

Cree

All ages

On-reserve
Aboriginals

All ages

James Bay,
Quebec

Cree

20 years +

Lac Simon,
Quebec

Algonquin

15 years +

River Desert,
Quebec

Algonquin

15 years +

Quebec

On-reserve
Aboriginals

All ages

Atlantic
Region

On-reserve
Aboriginals

All ages

Northwest
Territories

On-reserve
Aboriginals

All ages

Northwest
Territories

Inuit

All ages

Yukon
Territory

On-reserve
Aboriginals

All ages

Type of
Prevalence
(year)
Direct agestandardization
(1998)
Direct agestandardization
(1987)
Direct agestandardization
(1989)
Agestandardization
(1989)
Agestandardization
(1989)
Direct agestandardization
(1987)
Direct agestandardization
(1987)
Direct agestandardization
(1987)
Direct agestandardization
(1987)
Direct agestandardization
(1987)

Prevalence
Rate

Reference

10.3%

(Maberley et
ah, 2000)

7.6%

(Young,
Szathmary et
ah, 1990)
(Brassard et
ah, 1993)

6.6%

19%

(Delisle &
Ekoe, 1993)

9%

(Delisle &
Ekoe, 1993)

4.8%

(Young,
Szathmary et
ah, 1990)
(Young,
Szathmary et
a h ,1990)
(Young,
Szathmary et
ah, 1990)
(Young,
Szathmary et
ah, 1990)
(Young,
Szathmary et
ah, 1990)

8.7%

0.8%

0.4%

1.2%

As seen in Table 1, the prevalence rates of diabetes vary considerably between
different aboriginal groups across Canada. The Northwest Territories and the Yukon
Territory have the lowest prevalence rates of diabetes, while Ontario has some of the highest
rates. The overall prevalence rates of diabetes reported for British Columbia are quite low,
comparatively. However when just those over the age of 35 years are looked at, the

13

prevalence rates inerease substantially, indicating that the prevalence of diabetes is of
considerable eoneem among the aboriginal groups of British Columbia as well.
Prevalence Trends
There are three major trends seen in the diabetes prevalenee studies done with
aboriginal groups. The first major trend is that the prevalenee of diabetes increases with age
(Harris et al., 2003; Jorgensen et al., 2002). A study of Native Americans and Alaska
Natives found that the prevalence of diabetes increased from 0.2% in the under 20 years age
group to a peak of 22.8% in the greater than or equal to 65 years age group (Burrows &
Geiss, 2000). This trend is not exelusive to aboriginal groups, hut rather it is something that
is generally known about type 2 diabetes. However, aboriginal populations have a greater
proportion of their population being diagnosed with type 2 diabetes at a younger age than
their Caueasian eounterparts. In a study comparing the Indian and Caueasian populations of
Southwestern Ontario, those Indian individuals aged 25 to 34 years had a diabetes prevalenee
rate of 4.1%, while Caucasians in the same age group had a prevalenee rate of only 0.4%.
The difference between the two populations beeomes even more drastie at the 35 to 44 years
age group where the diabetes prevalence rates for the Indian population and Caueasian
population are 15.5% and 0.7%, respeetively (Evers et al., 1987).
The seeond major trend seen in the prevalence of diabetes among aboriginal
populations is that the prevalence is generally higher among women than men (Brassard et
al., 1993; Delisle & Ekoe, 1993; Harris et al., 2003; Johnson et al., 2002; Pioro et al., 1996;
Young, Szathmary et al., 1990). Burrows and Geiss (2000) noted that not only did women
have a higher prevalence of diabetes than men, but the prevalenee rate was aetually 1.4 times

14

higher. In a study of the Cree o f Western James Bay, 64% of the people with diabetes in
Moose Factory, Ontario were women (Maberley et ah, 2000).
The third and final major trend identified in many of the studies reviewed involves
the prevalence o f diabetes being related to the geographic location of the population. There
are two related facets to this trend, a north-south gradient and the effect of the relative
isolation and/or rurality of the community (Young, Szathmary et ah, 1990). Brassard,
Robinson and Lavallee (1993) examined the prevalence of diabetes in eight Cree villages
along James Bay in Quebec. The northernmost community had a low prevalence rate of
1.9% while the southernmost community had a considerably higher prevalenee rate of 9.0%.
The same north-south gradient was seen in Saskatchewan with only 1.7% of on-reserve First
Nations people living in the northern part of the province having diabetes, while 3.7% of the
First Nations population living in the south have the disease (Pioro et ah, 1996). In British
Columbia, the prevalenee of diabetes in the south, when considering all ages, is more than
one and a half times the prevalence in the north (3.0% and 1.8%, respectively) (Johnson et
ah, 2002). Another study done in British Columbia that broke down the geographic location
of the First Nations groups into four distinct locations (South Mainland, Vancouver Island,
Northwest and Northeast) again found the similar north-south gradient. It was also
discovered that while the northeast zone had the lowest prevalenee rate, that rate had also
increased three fold between 1987 to 1995, from 0.5% to 1.5% (Martin & Yidegiligne,
1998). Related to the issue of the north-south gradient, is the distinction between the
prevalence of diabetes in communities classified as urban, rural or remote. Brassard,
Robinson and Lavallee (1993) categorized the eight communities in their study based on
accessibility. Six communities where deemed to be remote given that they are primarily

15

accessible by air with very limited road aeeess. The other two communities were classified
as rural because they could be more easily accessed by roads. The study found that the
diabetes prevalenee rates were significantly higher in the two rural eommunities than the six
remote communities (8.8% and 3.4%, respectively). Young, Szathmary, Evers and Wheatley
(1990) defined urban as a community within 100 km by road of a Census Metropolitan Area,
rural as a community located further than 100 km from a Census Metropolitan Area and
accessible by road and/or rail, and remote as a community also further than 100 km from a
Census Metropolitan Area but only accessible by air or water. Generally, the prevalence rate
of diabetes was highest in the urban communities and lowest in the remote communities.

Theories Behind the High Prevalence of Type 2 Diabetes in Aboriginal Populations

Thrifty Genotype Theorv
The thrifty genotype theory is the most commonly cited theory when trying to reason
why the prevalence of type 2 diabetes is so high in some indigenous populations (Lieberman,
2003; Schulz, 1999). James V. Neel first introduced the theory in the early 1960’s in relation
to the high prevalence of diabetes seen in Native American populations. He postulated that
individuals with the ability to store energy when food was abundant and effectively use that
stored energy during times of food deprivation, had a selective advantage during the time
when the feast-and-famine cycle was common (Neel, 1962). Once the feast-and-famine
cycle was no longer an issue for these populations, this adaptive genotype that had once been
advantageous, was now detrimental to the health and well being of these people. It is
important to note that the thrifty genotype alone does not lead to the development of type 2
diabetes, but rather it is this genotype in conjunction with a change in lifestyle that seems to

16

lead to the development of type 2 diabetes. A study conducted by Schulz (1999) illustrates
the relative contribution of genetics and lifestyle to the development of type 2 diabetes
among the Pima Indians. The Pima Indians of Arizona have the highest prevalence rate of
type 2 diabetes reported, with reported rates ranging from 35 to 50% (Benyshek et al., 2001;
Wendorf, 1992; Zimmet, 1982). Schulz (1999) identified a genetically-related population of
Pima Indians living in an isolated community in Mexico where they continue to live a
predominantly traditional lifestyle. A comparison of these two groups was used to determine
the relative impact genetics and lifestyle have on the development of type 2 diabetes. The
Pima Indians of Arizona were forced to adopt a western lifestyle relatively quickly after
circumstances made continuing their traditional way of life, primarily associated with
agriculture, impossible. This change in lifestyle resulted in a diet that consisted of
considerably more fat and less dietary fiber than their traditional diet, and a severe reduction
in their activity level. In comparison, the Pima Indians of Mexico still maintain their
traditional lifestyle, which involves growing their own food and working labor-intensive
jobs. Their diet consists primarily of beans, com and wheat tortillas, potatoes and
inexpensive soups, and the consumption of animal products is rare. The Mexican Pima
Indians are incredibly physically active with the men working labouriously for approximately
48 hours per week, of which 40 hours are spent doing hard labour. The women, while rarely
working outside of the home, still spend approximately 91 hours per week doing labourious
household tasks, since there is no electricity or ranning water. These vast differences in
lifestyle have correlated with significant differences in the prevalence of type 2 diabetes. In
the study populations it was found that the prevalence of type 2 diabetes among the Arizona
Pima Indians was 38% while among the Mexican Pima Indians it was only 6%. Given the

17

genetic relatedness of the two populations, it stands to reason that lifestyle plays a significant
role in the development of type 2 diabetes.
Lieberman (2003), in her discussion of the dietary and modernizing influences on the
development of type 2 diabetes, detailed the nutritional transitions that many of the world’s
populations have gone through during the process of modernization. This transition has been
broadly divided into five phases. In the beginning, phase one, hunting and gathering
comprised most of the food collection. In phase two, diets exhibit less variety and there is
increasing incidence of starvation. This food deprivation led to the advent of agriculture.
With the agriculture came fewer periods of famine in phase three and once again the variety
of the diet increased with the addition of more fioiits, vegetables and animal products. It is in
phase four that the major lifestyle shift occurs, involving increased intake of animal fat,
sugar, and processed foods, and a reduction in the level of physical activity required to
sustain this lifestyle. This phase is currently being experienced by most modem. Western
civilizations and is at the point where many populations are developing an increased
prevalence of chronic conditions such as obesity and diabetes. The final phase in the
nutrition transition involves recognition of the problems created by phase four and an active,
conscious attempt to rectify them. This includes behavioural changes, a reduction in the
consumption of fat and processed foods, and an increased intake of fruits and vegetables. It
also involves the supplementation of physically active leisure activities to compensate for a
less physically demanding workload.
This change in lifestyle and the results of the study conducted on the Pima Indians in
Arizona and Mexico seem to support the thrifty genotype theory proposed by Neel all those
years ago. However, not everyone fully supports this theory. Benyshek, Martin and

18

Johnston (2001) in their paper entitled, A Reconsideration o f the Origins o f the Type 2
D iabetes Epidemic among N ative Americans and the Implications fo r Intervention Policy

point out several problems with the thrifty genotype theory and offer an alternative
hypothesis to the high prevalence of type 2 diabetes among Native Americans. Some of the
issues raised revolve around the genetic studies conducted in an attempt to prove the thrifty
genotype theory. There is concern surrounding methodological bias and the resulting
conclusions drawn from these studies. Secondly, the authors point out the inherent leap that
some researchers are making when generalizing the results of genetic studies on the very rare
forms of diabetes (e.g. maturity onset diabetes of the young) to the more common forms of
diabetes, namely type 1 and type 2 diabetes. Finally, Benyshek et al. state that there appears
to be little evidence that Native Americans went through regular periods of starvation, which
is a fundamental argument for the thrifty genotype theory.
An Alternate Hvpothesis: The Thriftv Phenotype
A study conducted in England was the basis for a new hypothesis about the origins of
the diabetes epidemic among certain populations (Hales et al., 1991). Rather than focusing
on the genotypic origin of diabetes, this hypothesis focuses on the phenotypic origin. The
aim of the study was to investigate whether there was a relationship between fetal growth,
specifically reduced fetal growth, and the development of type 2 diabetes. Using birth
weight and weight at age 1 year data for 468 men bom in east Hertfordshire, England
between 1920 and 1930, the association between these weights and the subsequent
development of type 2 diabetes later in life was examined. The authors discovered that adult
men with impaired glucose tolerance and type 2 diabetes were more likely to have a lower
birth weight and reduced weight gain during infancy. For example, of the men who had a

19

birth weight o f less than or equal to 5.5 pounds (or 2495 grams), whieh is the definition of
low birth weight, 30 pereent had impaired glucose tolerance and 10 percent had type 2
diabetes. Conversely, the percentage of men with impaired glucose tolerance and type 2
diabetes, whose birth weights were 9.5 pounds (or 4309 grams), was 4 percent and 9 percent,
respectively. From these results, the authors proposed that malnutrition during critical
periods of the pregnancy and, therefore, fetal growth and development, could prevent the
proper development of (3 cell function,

cells are a key component in the production of

insulin, as well as in the development of diabetes. It is thought that this change in the
development of (3 cells is the result of the fetus adapting to malnutrition in utero, and that
physiologic changes made during fetal development result in more permanent changes in the
body than those made during adulthood (Barker, 1999). While this adaptation increases the
survival of the fetus during times of malnutrition, it becomes a liability for the individual in
times of food abundance.
On the flip side of the low birth weight association is the association between high
birth weight, or maerosomia, and the subsequent development of type 2 diabetes.
Macrosomia is often the result of gestational diabetes or maternal diabetic status, which in
and of itself increases the risk of the offspring developing type 2 diabetes. Rieh-Edwards et
al. (1999), in their study of a cohort of American women, found a U-shaped association
between birth weight and type 2 diabetes, with both women of low birth weight and high
birth weight having an increased risk of developing type 2 diabetes. However, when the data
was adjusted to consider the presence of gestational diabetes or maternal diabetic status, the
U-shaped trend was replaced by an inverse relationship seen across all birth weights. A
study conducted with the Registered Indians in Saskatchewan noted the association between

20

high birth weight and type 2 diabetes, however, no significant association between low birth
weight and type 2 diabetes was found (Dyck, Klomp, & Tan, 2001). The authors recognized
that the high prevalence of maternal obesity, gestational diabetes and maternal diabetes could
account for the relationship between high birth weight and type 2 diabetes.
Both the thrifty genotype hypothesis and the thrifty phenotype hypothesis have their
strengths and weaknesses. While neither hypothesis has shown a definitive causal
relationship that can be explicitly proven, both have stimulated considerable research and
have a certain degree of plausibility. The quest to better understand the origins of the
diabetes epidemic seen in many populations has, as a result, focused attention on the problem
of diabetes and the need for more effective prevention and management strategies.

Research Questions

The purpose of this thesis is to answer the following questions about diabetes in the
Bella Coola Valley:
1. What is the prevalence of diabetes among the aboriginal and non-aboriginal people
living in the Bella Coola Valley?
2. How do these prevalence rates compare to the rates reported in the literature for other
aboriginal groups?
3. Are age, weight, body mass index, gender and aboriginal status significant risk
factors for developing diabetes?

21

Chapter Three
Methods
Research Design

This study is a retrospective population-based eross-sectional study. Cross-sectional
studies are the most common type of study done when examining the prevalence of disease.
A key feature to the cross-sectional study is that it measures both the exposure status and the
disease status at the same time. For this reason, one of the limitations to this type of study is
that it is generally not possible to determine whether or not the exposure preceded the
development of the disease or came as a result of the disease (Hennekens & Buring, 1987).
However, this limitation is diminished when looking at factors that are not changed by the
disease (Kelsey, Whittemore, Evans, & Thompson, 1996). In the context of this study, those
risk factors include age, gender and aboriginal status.

Study Community

The community chosen for study was the Bella Coola Valley. Located on the central
coast of British Columbia, the Bella Coola Valley is one of the most isolated rural
communities in British Columbia (Figure 1, 2). Nestled deep in the Coast Mountains, the
small town of Bella Coola is situated next to the Bella Coola River estuary, where the river
meets Burke Channel, the only port with access to inland British Columbia between
Vancouver and Prince Rupert. The main access to the Bella Coola Valley is by Highway 20,
an approximately 450 km stretch of highway that links the Bella Coola Valley with Williams
Lake in interior British Columbia. While this is a very scenic route to travel, it is also a
challenging drive particularly in the area just east of Bella Coola known as The Hill. This

22

area consists of a “steep, narrow road with sharp hairpin turns and two major switchbacks, as
the highway descends from the Chilcotin Plateau” (Bella Coola Valley Tourism, 2003). In
addition to road access, Bella Coola can also be accessed by air and water. There is a daily
flight between Bella Coola and Vancouver all year round, with an additional flight offered
during the summer. BC Ferries operates a Discovery Coast Passage route, which links Bella
Coola to Port Hardy on Vancouver Island (Bella Coola Valley Tourism, 2003).
The economy of the Bella Coola Valley is primarily dependent upon natural
resources. Commercial salmon fishing and forestry have been the major contributors to the
local economy; however, with the cutbacks seen province-wide in both of these areas,
tourism is becoming an increasingly important component (Central Coast Economic
Development Commission, 1990). The unemployment rate in 2002 for the Central Coast
Regional District, of which Bella Coola is the major centre, was 8.0%, substantially higher
than the provincial average of only 3.6% (BC Stats, 2002).

23

\ / )

I

n

\./
I

\
\

I \ NA li A

/

BRITISH
COLUMBIA
1/

< :!(:)i,O M ]ii]E :-

BRITANNIQUE

I

f:
%

" 'Î Ï
LEGEND/ LÉGENDE
_______ In te rn a tio n a l b o u n d a ry /
F r o n h è r e i n t e r n a t io n a l e

\

__ ___ Provincial boundary /

Limite provinciale

1

(,* Bella Coola
' Williams
Lake .

V
V.

X .

'n
%

Scmk / Échelle
km L_±.

*

100

m

.Vancouver

KO

Figure 1. Map of British Columbia highlighting the locations of Bella Coola, Williams
Lake and Vancouver.

24

Kilometers
\ MORESBY
^IS L A N D
Ootsa

««iN»

Mf»'

t

K*.

rr^ W r

Figure 2. Detailed map of the Bella Coola Valley,

25

Bella Coola Valley Population

The population of the Bella Coola Valley is approximately 2,260 people according to
the 2001 census data. To determine the population of the Bella Coola Valley, and not the
larger Central Coast Regional District, the populations of individual Regional District
Electoral Areas for the Central Coast (Central Coast C, Central Coast D, and Central Coast
E), in addition to the Bella Coola Indian Reserve, were aggregated to get an approximation of
the population of the Bella Coola Valley (Figure 3). The age and gender breakdown of the
total population of the Bella Coola Valley is provided in Table 2.

Table 2. Age and Gender Breakdown of the Bella Coola Valley based on the 2001
Census (Statistics Canada, 2001).
Age Group
Age 0 - 4
Age 5 - 1 4
Age 1 5 -1 9
Age 20 - 24
Age 25 - 44
Age 45 - 54
Age 55 - 64
Age 65 - 74
Age 75 - 84
Age 85 and over
Total

Male
90
185
85
60
355
190
110
60
25
10
1170

Female
65
185
100
50
340
155
90
70
30
5
1090

26

Central Coast Regional District
BULKLEY-NECHAKO

KITIMAT-STIKINE,

C A R IB O O

MOUNT WADDINQTON

Produced by BC Slals

Figure 3. Map of the Central Coast Regional District Electoral Areas. Areas C, D, and
E were used to calculate the census population for the Bella Coola Valley.

A significant portion of the population of the Bella Coola Valley is made up of
aboriginal people, approximately 44% according to the 2001 census. The majority of these
aboriginal people live on the Bella Coola Indian Reserve as indicated in Table 3.

27

Table 3. Number of People Living On and O ff Reserve in the Bella Coola Valley, 2001
Census (Statistics Canada, 2001).

Living Off Reserve

Living On Reserve

First Nations

135

855

Non-First Nations

1240

55

Total

1375

910

Most of the aboriginal people living in the Bella Coola Valley are of Nuxalk deeent. The
Nuxalk are part of the Salish language group and have traditionally lived within the Bella
Coola Valley and surrounding central coast areas (Thommasen, Loewen, & Mclnnes, 1995).
They were a settled tribe; living in long houses, constructed of wood harvested locally, with
often more than one family residing in each of these houses. The villages were built along
the Bella Coola River, allowing easy access to the river itself. The Nuxalk were primarily
hunters, fishermen, and traders before they came in contact with the Europeans. They hunted
deer, mountain goat, moose, grizzly bear, seal, and duck. The large eulachon fish runs
provided the sustenance of their trading, as the grease produced from these fish was traded
with inland tribes (Central Coast Economic Development Commission, 1990).
Contact with Europeans began in the late 18*^ century with the arrival of the
explorers. By the early 19*'’ century, the Hudson’s Bay Company had built forts in the area
and had established a regular trade route. This contact led to a change in the Nuxalk culture,
with the people now abandoning some of their traditional activities in order to stay closer to
the established forts. This regular contact with foreign people, and thereby foreign diseases.

28

made the Nuxalk population very vulnerable. In fact, the smallpox epidemic in 1862 nearly
destroyed the native population (Central Coast Economic Development Commission, 1990).

Health Services in the Bella Coola Valley

The only hospital and clinic in the Bella Coola Valley is located within the town of
Bella Coola itself. Bella Coola General Hospital, operated by the United Church Health
Services, delivers a wide range of health services including surgery and medicine, on both an
outpatient and inpatient basis, and 24 hour emergency coverage (Thommasen, Newberry, &
Watt, 1999; Vancouver Coastal Health Authority, 2003). In addition, it also has onsite
laboratory, x-ray, and physiotherapy services. The hospital and adjacent clinic are served by
three physicians and one family practice resident physician (Bella Coola General Hospital,
n.d.). On average these physicians see approximately 8,000 clinic visits, 2,500 emergency
room visits, and 400 hospital admissions (Gobrial, Mekael, Anderson, Ayers, & Thommasen,
2002y
The isolated nature o f Bella Coola, and thereby the Bella Coola General Hospital and
adjacent clinic, results in almost everyone who lives in the Bella Coola Valley having either a
clinic chart or emergency room record. This makes Bella Coola an ideal location to do this
type of population-based study.

Data Collection - The Bella Coola Valley Clinic Population

A retrospective chart review was conducted by Dr. Harvey Thommasen, a Family
Physician at the Bella Coola Medical Clinic, of all of the clinic charts. An estimated 560
hours were spent reviewing the approximately 2,700 clinic charts. The criterion for inclusion

29

in this study population was a current address loeated within the Bella Coola Valley.
Therefore, residents of Anahim Lake, Nimpo Lake, Ocean Falls and Bella Bella were
excluded from the study population.
After exeluding those clinic charts of people who did not eurrently live within the
Bella Coola Valley, 2,377 patients made up the study population. The information collected
during the ehart review, in addition to address, included age, gender, weight, body mass
index, aboriginal status, and presence or absence of diabetes. Age was caleulated as of May
2001 based on the patient’s birth date. The weight recorded was the most recent on file and
the body mass index was calculated for those patients who had both a weight and height
value on file. The presence of diabetes was based on a physician diagnosis of the disease,
which, in turn, was based on the 1998 clinical practice guidelines for the management of
diabetes in Canada (Meltzer et al., 1998).
Aboriginal status for the study population was determined from multiple sourees.
Aboriginal status for Nuxalk Band members was determined from the 1920, 1979, 1989, and
2001 Nuxalk Band lists, as well as the birth and death vital statisties for the Nuxalk Band
members. In consultation with Nuxalk elders. Dr. Harvey Thommasen constructed a
comprehensive geneology of the Nuxalk people in the 1990’s that was also used to assign
aboriginal status to the Nuxalk patients in the study population. There are, however,
aboriginal people living in the Bella Coola Valley who are not of Nuxalk decent. These
people were identified from a review of their charts; from their response to a survey question
asking about aboriginal status; or by asking them directly about their aboriginal ancestry.
After aboriginal status was assigned to each patient in the study population by Dr. Harvey
Thommasen; a long-time resident and Medical Clinic staff member then reviewed, and

30

verified or questioned, the eorreetness of the aboriginal status assigned aeeording to the best
of that person’s knowledge. There were only a few alterations made after consultation with
the Medical Clinic staff person. Approximately 47% of the people living in the Bella Coola
Valley are of aboriginal decent, according to the aboriginal status assigned to the study
population. The 2001 census estimates that approximately 44% of the population of the
Bella Coola Valley is of aboriginal deeent. Table 4 summarizes the age and sex breakdown
of the Bella Coola Valley study population.

Table 4. Age and Gender Breakdown of the Bella Coola Valley Study Population

Age (years)
0-4
5-14
15-19
20-24
25-44
45-54
55-64
65-74
75-84
>85
Total

Total Population
155
367
215
147
686
369
220
141
62
15
2377

Male
85
191
103
72
352
202
115
72
25
7
1224

Female
70
176
112
75
334
167
105
69
37
8
1153

According to the 2001 census data, there are 2,260 people living in the Bella Coola
Valley. The estimation o f the Bella Coola Valley population, based on the clinic chart
review, is 2,377, or 105% of the 2001 census population. For this reason, we are confident
that we have captured the entire Bella Coola Valley population. Table 5 compares the age
and sex breakdown of the census population to the study population.

31

Table 5. Comparison of 2001 Census Population and Study Population, Broken Down
by Age and Gender, for the Bella Coola Valley.

Male
Age (years)
0-4
5-14
15-19
20-24
25-44
45-54
55-64
65-74
75-84
>85
Total

2001 Census
Population
90
185
85
60
355
190
110
60
25
10
1170

Study
Population
85
191
103
72
352
202
115
72
25
7
1224

Female
2001 Census
Study
Population
Population
65
70
185
176
100
112
50
75
340
334
155
167
90
105
70
69
30
37
5
8
1090
1153

All of the data was entered into an Excel spreadsheet by Dr. Harvey Thommasen.
Once he had confirmed that the data transfer was accurate, all of the identifying pieces of
information were removed and replaced with identification numbers. This process was to
ensure the confidentiality of the people of the Bella Coola Valley. It was at this point that the
dataset became available to be used for analysis in this study.

Ethics Approval

Ethics approval to collect the data related to this study was obtained from the
Research Ethics Committee at the University of British Columbia by Dr Harvey Thommasen.
Ethics approval to use the data in this particular thesis was obtained from the Research Ethics
Board at the University o f Northern British Columbia (Appendix I).

32

Data Analysis

Data analysis was performed using Statistical Package for the Social Sciences (SPSS)
for Windows. All analyses were conducted on the total adult study population (> 18 years)
and for aboriginals and non-aboriginals separately, where appropriate. To answer the
research questions, the following analysis strategies were utilized:
What is the prevalence o f diabetes among the aboriginal and non-aboriginal people
living in the Bella Coola Valley?

The crude prevalence rates of diabetes were calculated through simple frequency
tables. The age-adjusted prevalence rates were calculated using the direct method of
adjustment with the standard population being a distribution of the aboriginal and non­
aboriginal populations combined (Hennekens & Buring, 1987). Prevalence rates were
broken down by gender, age groups, weight and body mass index categories.
How do these prevalence rates compare to the rates reported in the literature fo r
other aboriginal groups?

The prevalence rates determined for the Bella Coola Valley aboriginal population
were compared to the rates reported for other aboriginal populations with particular attention
paid to the prevalence trends observed in the literature and how they compare to the trends
observed in the Bella Coola Valley study population.
Are age, weight, body mass index, gender and aboriginal status significant risk
factors fo r developing diabetes?

For the analysis of statistical significance the dependent variable is diabetic status as a
dichotomous variable. The difference in prevalence rates based on gender and aboriginal

33

status were evaluated using Pearson’s chi-squared. The differences based on age, weight,
and body mass index, all continuous variables, were evaluated using ANOVA.
Odds ratios were calculated using logistic regression to determine the predictive
power of age, weight, body mass index, gender, and aboriginal status. A univariate analysis
was performed on each risk factor. Subsequently, a multivariate analysis was performed.
Both weight and body mass index could not be included in the multivariate analysis because
they are too highly correlated. Given the large number of body mass index values that were
missing from the dataset, weight was chosen for inclusion in the multivariate analysis.
For all of the analyses significance was set at p < 0.05.

34

Chapter Four
Results

The characteristics of the study population, divided into aboriginal and non-aboriginal
subsets, are summarized in Table 6.

Table 6. Summary of the Characteristics of the Adult Study Population

Characteristic
Gender
Male
Female
Age (years)
1 8 -2 4 .9
2 5 -3 4 .9
3 5 -4 4 .9
45 - 54.9
55 - 64.9
^ 65
Weight (kg)^
<60
6 0 -6 9
7 0 -7 9
8 0 -8 9
9 0 -9 9
1 0 0 -1 0 9
> 110
Body Mass Index (kg/m^)’’
<25
2 5 -2 & 9
3 0 -3 4 .9
3 5 - 3 9 .9
>40
^

Aboriginal
N(% )

Non-aboriginal
N(% )

366(50.(%
353 (49T)

518(51.5)
488(48.5)

p Value
0.810

0.000
140(19.5)
170(23.60
156 (21.7)
129(17.9)
66 (9.2)
58(8.1)

92 (9.1)
143 (14.2)
217(21.60
240(23.90
154(15.3)
160(15.9)

0.182
65 (9.0)
105 (14.6)
145 (20.2)
148(20.60
99 (13.8)
62 (8.6)
58 (8.1)

98 (9.7)
178(17.7)
200(19.90
186 (1&5)
121 (12.0)
75 (7.5)
55 (5.5)
0.000

83 (11.5)
139(19.3)
107(14.9)
87(12.1)
48 (6.7)

203 (20.20
253 (25.1)
126(12.5)
52(5.20
23 (2.3)

W eight data is m issing from 130 (7.5) patients [37 (5.1) aboriginal; 93 (9.2) non-aboriginal]

^ B ody mass index data is m issing from 604 (35.0) patients [255 (35.5) aboriginal; 349 (34.7) non-aboriginal]

35

There is a similar gender distribution in both the aboriginal and non-aboriginal
populations, with both populations having nearly a IT ratio of males to females. The non­
aboriginal population is significantly older than the aboriginal population (p < 0.001), with
mean ages o f 40.28 ± 21.00 years and 29.09 ± 19.51 years, respectively. The non-aboriginal
population also has a higher mean weight (70.87 ± 26.31 kg vs. 65.77 ± 30.45 kg) than the
aboriginal population, however this difference was not significant. The non-aboriginal
population also bad a lower mean body mass index (25.92 ± 6.33 kg/m^ vs. 27.25 ± 8.02
kg/m^) than the aboriginal population, and this difference was found to be significant
(p < 0.001). This indicates that the aboriginal population is potentially more overweight than
the non-aboriginal population.
The overall crude prevalence of type 2 diabetes was higher in the aboriginal adult
population than in the non-aboriginal adult population (9.7% vs. 5.7%). Given the
significant differences in the age distributions of the aboriginal and non-aboriginal
populations, and the important role that age plays in the prevalence of diabetes, it was
important to calculate the age-adjusted prevalence rates for each of the populations. Details
of these calculations can be found in Appendix II. The age-adjusted prevalence rates of
diabetes for the aboriginal and non-aboriginal populations were 12.5% and 4.8%,
respectively. Figures 4 and 5 illustrate the trends in type 2 diabetes prevalence based on age
and gender. For each graph, the prevalences in both the aboriginal and non-aboriginal
populations are compared. The relationship between aboriginal ancestry and diabetic status
was found to be statistically significant (%^ = 10.183; p = 0.001).

36

45.0 n
40.0

40.0 34.3

35.0
30.0
O)

2 25.0 a

ü Aboriginal
B Non-aboriginal

2 20.0
0)
Q.
15.0
10.0

5.0

0.00.0 ^0.0
0.0 -

18-24.9 25-34.9 35-44.9 45-54.9 55-64.9 65+

Total

Age (years)

Figure 4 . The prevalence of type 2 diabetes in aboriginal and non-aboriginal males
based on age.

33.3

35.0
29.0

30.0

o 25.0
O)
IQ
.

c

20.0

ü Aboriginal

0)
2 15.0
01
Q.

B Non-aboriginal
10.2

10.0
4.6

5.0

0.00.0
0.0

^'^0.0

I

18-24.9 25-34.9 35-44.9 45-54.9 55-64.9 65+

Total

Age (years)

Figure 5. The prevalence of type 2 diabetes in aboriginal and non-aboriginal females
based on age.

37

Aside from the aboriginal population having a eonsistently higher prevalenee rate of
diabetes, a trend of the rate increasing with increasing age can also be clearly seen in both
males and females. The overall prevalence of diabetes for non-aboriginal males is slightly
higher than for non-aboriginal females (6.2% vs. 5.1%). The same is true for the aboriginal
population overall (males = 10.1%; females = 9.3%), though aboriginal females aged 35 to
44.9 years have a prevalenee rate substantially higher than aboriginal males in the same age
group (9.7% vs. 3.6%). Despite the generally higher prevalence rates observed in males,
there was no significant relationship found between gender and diabetic status. There was,
however, a significant relationship observed between age and diabetic status, in both the
aboriginal (F = 25.98; p < 0.001) and non-aboriginal populations (F = 13.41; p < 0.001).
Not surprisingly, based on the literature, weight was found to be significantly related
to diabetic status (aboriginal: F = 7.70; p < 0.001, non-aboriginal: F = 6.34; p < 0.001). As
seen in Figure 6, the prevalence rate of diabetes generally increases with increasing weight.

38

35.0
30.0
25.0
@
D)
jS 20.0
c
0)
2 15.0
o>

■ Aboriginal
H Non-aboriginal

Q.

10.8

10.0
: 5.4

5.0

a I

3.1

2.S

<60

60-69 70-79 80-89 90-99

0.0

100109

110+

W eight (kg)

Figure 6. Prevalence rate of type 2 diabetes among aboriginal and non-aboriginal
adults based on weight (kg).

With the exception of the 70 - 79 kg weight category, the aboriginal population had a higher
prevalence rate of type 2 diabetes based on weight, with the difference being extremely
apparent in the greater than 110 kg weight category (31.0% vs. 20.0%).
Body mass index is another measure used to evaluate a person’s weight. This
measure, however, takes a person’s height into account and is often considered a more
accurate evaluator of obesity. The calculation for body mass index (BMl) is:
BM I =

Weight(kg)
Height {m )

Figure 7 illustrates the distribution of prevalence rates of type 2 diabetes in both the
aboriginal and non-aboriginal populations based on body mass index.

39

50.0

45.8

45.0 4
40.0
35.0
« 30.0
a,

B A boriginal
■ Non-aboriginal

< 25

25-29.9

30-34.9

35-39.9

40+

Body Mass Index (kg/m^)

Figure 7. Prevalence rates of type 2 diabetes among aboriginal and non-aboriginal
adults based on body mass index (kg/m^).

In both the aboriginal and non-aboriginal populations, the prevalenee of diabetes increases
with increasing body mass index. The rate of increase seen is more substantial in the
aboriginal population with the prevalenee rate approximately doubling with each increase in
body mass index category. The increase seen in the non-aboriginal population is more
gradual. Despite these slightly different trends, body mass index is significantly related to
diabetic status for both populations (aboriginal: F = 15.3; p < 0.001, non-aboriginal: F = 8.02;
p < 0.001).
Logistic regression was used to provide a more detailed analysis of the increased risk
of developing diabetes associated with age, weight, body mass index, gender, and aboriginal
status. While the previous analyses elude to a significant relationship between the factors of
age, weight, body mass index, and aboriginal status and diabetic status, logistic regression
examines the contribution each factor makes to the overall risk of developing diabetes. Table
7 outlines the odds ratios and associated p values for gender, age, weight, body mass index.
40

and aboriginal status. The results of both the univariate and multivariate analyses are
included. For the multivariate model, both weight and body mass index could not be
included in the same model due to the fact that these factors are highly correlated. Therefore,
a choice had to be made with regards to which factor to include. Weight was chosen for the
multivariate model because there was less data missing for this factor in the dataset, therefore
the model using weight was a more accurate reflection of the study population.

41

Table 7. Odds Ratios (OR), p values and 95% Confidence Intervals (Cl) for Diabetes
from Univariate and Multivariate Logistic Regression Models
Univariate Analysis"
OR
95% C l

Risk Factor
Gender
Male
Female

Reference
0.9

Age (years)
1 8 -3 4 .9
3 5 -4 4 .9
45 - 54.9
5 5 -6 4 .9 .9
> 65

Reference
9.8
21.4
622
68.6

Weight (kg)
<60
60 - 69
7 0 -7 9
8 0 -8 9
9 0 -9 9
1 0 0 -1 0 9
>110

Reference
4.3
14.1
182
30j
324
782

Body Mass
Index (kg/m^)
<25
2 5 -2 ^ 9
3 0 -3 4 .9
3 5 -3 9 .9
> 40

Reference
7.0
12.1
21.2
56.1

Aboriginal
Status
Non­
aboriginal
Aboriginal

p-value

Multivariate Analysis'*
OR
95% C l

p-value

0.458

Reference
1.3

—
0.9-2.0

—
&188

2.2-43.7
5.1-90.7
14.9-259.6
16.5-286.0

0.003
0.000
0.000
0.000

Reference
10.8
24.0
7&0
107.4

2.4-48.7
5.6 -103.2
18.2-333.4
25.0-460.9

0.002
0.000
0.000
0.000

1.1 - 17.3
4.2-47.7
5.6-62.4
9.1-101.8
9.3-112.1
23.4 - 262.6

0.040
0.000
0.000
0.000
0.000
0.000

Reference
2.0
4.9
6.6
9.3
11.3
31.4

0.4-10.3
1.1-22.1
1.5-29.3
2.1-41.9
2.4 - 52.9
6.8-144.8

0.413
0.037
0.014
0.004
0.002
0.000

3.1 - 16.2
5.2-28.0
9.0-49.7
23.2- 135.4

0.000
0.000
0.000
0.000

Reference
N/A
N/A
N/A
N/A

—

—

—

—

—

—

—

—

Reference

—

-

Reference

—

—

1.4

1.0-2.0

0.053

3.0

2.0-4.6

0.000

—

—

0.6-1.2

5 univariate m odels with one risk factor in each logistic regression model.
1 multivariate model with 4 risk factors (gender, age, w eight and aboriginal status) in the logistic regression
model.

42

For the purposes of the logistie regression, the age groups of 1 8 -2 4 .9 years and 25 34.9 years had to be collapsed into one age group (18 - 34.9 years) due to the lack of eases of
diabetes in the 18 - 24.9 years group. The lack of cases in this group makes it invalid as a
reference group, and therefore makes the calculation of odds ratios impossible. The addition
of the 25 - 34.9 years age group, which has two cases of diabetes, makes the combined group
a valid reference group. For weight, the less than 60 kg group was chosen as the reference
group because it is reasonable to assume that an adult weighing less than 60 kg should not be
overweight and therefore, should be at a low risk of developing diabetes based on weight. A
similar rationale was used in choosing the less than 25 kg/m^ group as a reference for body
mass index. According to the World Health Organization, a body mass index of less than 25
kg/m^ is considered to be normal, while a body mass index of 30 kg/m^ or greater is
considered to be obese (Puska et al., 2003).
The results of the univariate model indicate that age, weight, and body mass index are
all significant risk factors for developing diabetes. An increase in the value of any of these
factors corresponds with an increase in the respective odd ratios. For example, the odds ratio
for ages 35 through 44.9 years is 9.8, while the odds ratio for ages 55 through 64.9 is 62.2, an
increase of more than 6 fold. Similar to the results found in the chi-squared analysis, gender
was not found to be a significant risk factor for the development of diabetes. Aboriginal
status, in the univariate model, was right on the cusp of being significant (p = 0.053) based
on the set significance level of p < 0.05.
When comparing the univariate model to the multivariate model, the odds ratios are
greater in the multivariate model than the univariate model for all of the risk factors with the
exception of weight. Not only do the odds ratios decrease for weight in the multivariate

43

model, but the 60 - 69 kg weight eategory is also not significant. In the univariate model, all
of the weight categories are significant. In the multivariate model, aboriginal status becomes
significant (p = 0.000) with the odds ratio doubling from the univariate model, from 1.4 to
3.0. Therefore, it can be interpreted that after controlling for all other factors, those with
aboriginal ancestry are three times more likely to develop diabetes.

44

Chapter Five
Discussion

The findings of this study support what was previously reported in the literature
regarding the prevalence of diabetes among aboriginal groups as compared to the general
population. In the Bella Coola Valley study population, the crude prevalence of diabetes in
the aboriginal population is more than one and half times that of the non-aboriginal
population (9.7% vs. 5.7%). However, when the differences in the age distribution of the
aboriginal and non-aboriginal populations are considered, and the prevalence rates are
adjusted, the prevalence of diabetes in the aboriginal population is more than two and half
times that of the non-aboriginal population (12.5% vs. 4.8%). This indicates that diabetes is
of substantial concern to the aboriginal people of the Bella Coola Valley. The age-adjusted
prevalence rate of diabetes for the non-aboriginal population is the same as what has been
reported for the general population of Canada (Health Canada, 2002).
The age-adjusted prevalence rate of diabetes among the aboriginal people in Bella
Coola is substantially higher than what has been previously reported for First Nations groups
in British Columbia. Young, Szathmary, Evers and Wheatley (1990) did a survey of the
prevalence of diabetes in First Nations groups across Canada and reported that the ageadjusted prevalence rate of diabetes among the First Nations population of British Columbia
was 1.6%. While this survey is out-of-date now, another study of the prevalence of diabetes
in British Columbia conducted in the late 1990’s indicated that the crude prevalence rate of
diabetes in First Nations people of all ages was 2.6%, still considerably lower than the crude
prevalence rate found for the Bella Coola Valley aboriginal population. It is important to
note, when comparing prevalence rates tfom different studies, the type of prevalence

45

reported, crude or age-adjusted, and the age groups included. For example, a prevalence rate
that includes only the adult population will usually be higher than a prevalence rate that
includes all ages, because cases o f type 2 diabetes are more likely to be found in adults than
children.
In the review of the literature conducted prior to this study, three major trends in the
prevalence rates of diabetes among aboriginal people were identified. The first trend
identified was that the prevalence of diabetes increases with age. This trend was also seen in
the Bella Coola Valley study population, for both the aboriginal and non-aboriginal
populations (Figure 4 and 5). While both populations experienced an increase in prevalence
rate with an increase in age, the amount of increase was much higher for the aboriginal
population, from 0% in the 18 - 24.9 years category to 34.5% in the 65 years and greater
category. A similar trend was found in a study comparing Indians to Caucasians in
southwestern Ontario. In this study the prevalence of diabetes in the Indian population
increased from 1.0% in the 15 - 24 years category to 43.8% in the 75 years and greater
category, while the prevalence in the Caucasian population only increased from 0.5% to
13.2% in the same age groups (Evers et ah, 1987).
In addition to the prevalence of diabetes increasing with age, another age related trend
is seen specifically within the aboriginal populations. Aboriginal populations are
experiencing a younger age of onset of diabetes as compared to the non-aboriginal
populations. In the Bella Coola Valley study population, there were no cases of diabetes
identified in the non-aboriginal population prior to the age of 40 years. However, in the
aboriginal population a number of cases were identified in the 25 - 34.9 years age category.
A younger age of onset has been observed in other studies of diabetes and aboriginal

46

populations (Evers et al., 1987; Harris et al., 2003; Young, Reading, Elias, & O'Neil, 2000).
The results of this study are particularly concerning with regards to the younger age of onset
observed among aboriginal women. When the prevalence rate of diabetes is isolated to those
aboriginals aged 25 - 39.9 years, there is a substantial difference between the sexes. While
the prevalence rate for aboriginal men in this category was 0.8%, the prevalence rate of
diabetes in aboriginal women for the same age group is 4.8%. The rate among aboriginal
women in this age group is of concern because these are the primary childbearing years for a
woman, and there is increasing evidence of a link between maternal diabetes and the
subsequent development of diabetes in her offspring.
A study of the Pima Indians in Arizona found that 10 - 15% of pregnancies were
complicated by diabetes (Dabelea et al., 2000). Infant macrosomia is often the result of
diabetic pregnancies and can lead to such complications as the need for a caesarian section,
birth trauma and injury, and infant morbidity (Rodrigues, Robinson, Kramer, & GrayDonald, 2000). Some of the consequences to the children of maternal diabetics include
increased risk of obesity and the development of type 2 diabetes. In the study of the Pima
Indians it was found that approximately one-third of the cases of type 2 diabetes in children
may have been the result of exposure to a diabetic intrauterine environment (Odds Ratio =
10.4). The link between maternal diabetes and the subsequent development of diabetes in
offspring creates a vicious cycle in which diabetes can be perpetuated through the
generations. A woman with diabetes is at risk for a complicated pregnancy and delivery.
The infant, once bom, is at an increased risk of being obese and developing type 2 diabetes.
If that infant is a girl, she has the risk of developing type 2 diabetes before or during her
childbearing years and, therefore, starting the whole cycle over again (Dabelea et al., 2000).

47

The second trend observed in the previous studies conducted on the prevalence of
diabetes among aboriginal people is that the prevalence of diabetes is generally higher among
women than men. This is a trend that has been well documented in the literature (Brassard et
al., 1993; Burrows & Geiss, 2000; Delisle & Ekoe, 1993; Harris et al., 2003; Young,
Szathmary et al., 1990), yet was not observed exactly in this study. For ages 25 - 44.9 years
women did have a higher prevalence of diabetes, however from age 45 years and older,
aboriginal men had the higher prevalence rates. Despite these trends the relationship
between gender and diabetic status was not found to be significant in this study.
The final trend seen in the literature was that the prevalence of diabetes was related to
the geographic location o f the population. Of particular relevance to this study is the
relationship between the remoteness of the location and the prevalence of diabetes. Other
studies have found that the prevalence of diabetes is highest in urban communities and lowest
in remote communities (Brassard et al., 1993; Young, Szathmary et al., 1990). While this
study did not set out to specifically compare two geographic locations, it does stand to
reason, given the isolated nature of the Bella Coola Valley, that the prevalence of diabetes
should have been low. Compared to the remote communities studied in northern Quebec, the
crude prevalence of diabetes in the First Nations people of the Bella Coola Valley is more
than two and half times that of the James Bay Cree (9.7% vs. 3.4%) (Brassard et al., 1993).
This indicates that type 2 diabetes is a significant factor for the aboriginal people of the Bella
Coola Valley, and that the isolation of where they live may not have the same protective
effect that is seen in other isolated aboriginal groups.
The prevalence of diabetes among the non-aboriginal populations has not been as
extensively studied. Rather surveys of the general population are used to determine the

48

prevalence of diabetes and other diseases and/or conditions. For example, the National
Population Health Survey in 1998/99 determined that the prevalence of diabetes in the
general population of Canada was 4.8% (Health Canada, 2002). While this is exact same
age-adjusted prevalence rate that was found in the non-aboriginal population of the Bella
Coola Valley, the crude prevalence rate in the non-aboriginal population was slightly higher
(5.7%). Since the age-adjusted prevalence rates were adjusted to compensate for the age
distribution differences between the aboriginal and non-aboriginal populations of the Bella
Coola Valley, the crude prevalence rate is probably the better rate to use to compare the non­
aboriginal population to other general populations. The National Population Health Survey
in 1998/99 also found that men had a slightly higher prevalence of diabetes than women
(5.0% vs. 4.6%) (Health Canada, 2002). This trend was also mirrored in the Bella Coola
Valley non-aboriginal population with men having a diabetes prevalence rate of 6.2% while
females had a rate of 5.1%. Another study of Canadian adults, age 40 years and older, found
that men were 1.6 times more likely to have diabetes than women (Choi & Shi, 2001).
Several potential risk factors for diabetes were recorded for the Bella Coola Valley
study population, including age, gender, weight, body mass index and aboriginal status.
While this is not an exhaustive list of potential risk factors for diabetes, is does include many
of the major risk factors identified in the literature, and are ones that can be easily abstracted
from medical charts. Since the Bella Coola Valley study population included both aboriginal
people and non-aboriginal people, it was possible to compare the two groups in terms of their
prevalence of diabetes. In addition to this comparison, it was also possible to determine if
there was a significant relationship between aboriginal ancestry and diabetic status,
something that has not been done explicitly in the literature. Two Canadian studies were

49

found that compared aboriginal people to non-aboriginal people with regard to the prevalence
of diabetes (Evers et al., 1987; Pioro et al., 1996). While both of these studies compared the
prevalence rates o f diabetes, and other demographic and/or risk factor variables, neither study
specifically examined the relationship between aboriginal ancestry and diabetic status
looking for statistical significance. In the Bella Coola Valley study, both chi-squared
analysis and logistic regression were used to investigate the relationship between aboriginal
ancestry and diabetic status. It was determined through the chi-squared analysis that there is
indeed a significant relationship between aboriginal ancestry and diabetic status

= 10.183;

p = 0.001). Two different logistic regression models were used to explore the contribution of
aboriginal status to the risk of developing diabetes. The first model was a univariate model
that just looked at aboriginal status and diabetic status, without controlling for any other
factors. The results of this analysis indicated that aboriginals were 1.4 times more likely to
develop diabetes, though this relationship was not quite significant (p = 0.053) based on our
set significance level of p < 0.05. The second model, a multivariate model, controlled for
age, gender, and weight when examining the relationship between aboriginal status and
diabetic status. The resulting odds ratio indicated that, after controlling for these other
confounding factors, aboriginals are three times more likely to develop diabetes than non­
aboriginals. This implies that, in addition to aboriginals just having a higher prevalence of
diabetes, aboriginal ancestry is a significant risk factor for diabetes above and beyond age,
gender and weight.
A well known fact about type 2 diabetes is that the prevalence of the disease increases
with age. This trend was seen in the Bella Coola Valley study for both the aboriginal and
non-aboriginal populations. Therefore, it was not surprising to discover that age is a

50

significant risk factor for the development of diabetes. This determination was made through
both the ANOVA and logistic regression analyses. Using the 18 -34.9 years category as a
reference, the univariate logistic regression indicated that the risk of developing diabetes
increases substantially with age. The increased risk is even more substantial when gender,
weight and aboriginal status have been controlled for, as they were in the multivariate model.
For example, the odds ratio for the 35 - 44.9 years age group increases from 9.8 to 10.8 when
moving from the univariate to multivariate model. There is an even more drastic increase
when looking at the 65 years and older age group, with the odds ratio going from 68.6 to
107.4. Choi and Shi (2001) found that the odds of developing diabetes increased by 9% per
year of increase in age and Fletcher et al. (2002) noted that the fact that type 2 diabetes was
previously referred to as adult-onset diabetes, which emphasizes the relationship between age
and the risk of developing diabetes. This strong relationship between age and the risk for
developing diabetes has been widely established and accepted within the literature and
medical community, and has been further supported by this study (Jorgensen et al., 2002;
Levin, Mayer-Davis, Ainsworth, Addy, & Wheeler, 2001; Young, Moffatt, & Ling, 1988).
Another commonly examined risk factor in association with diabetes is obesity.
There are a variety of measurements used to evaluate obesity, including weight, body mass
index, waist-to-hip ratios, and percentage of body fat. The most eommonly reported measure
of obesity is the body mass index (Jorgensen et al., 2002; Motala, Pirie, Gouws, Amod, &
Omar, 2003; Young & Harris, 1994). While not a perfect measure, body mass index more
accurately reflects the presence or absence of obesity because it measures weight in relation
to height. Weight alone ean be a very subjective measure because its health impact will vary
depending on how that weight is distributed. For example, a person who weights 70 kg and

51

is 1.5 m tall is far more overweight (BMI ==31 kg/m^) than someone who weighs the same
but is 1.8 m tall (BMI = 21 kg/m^). In fact, according to the World Health Organization’s
definition of obesity, the person who weights 70 kg and is 1.5 m tall would be considered
obese (BMI > 30 kg/m^) while the person who weighs 70 kg and is 1.8 m tall would be
considered as having a normal weight (BMI = 18.5 - 25 kg/m^) (Puska et al., 2003). In the
Bella Coola Valley study population this difference is illustrated by the fact that the mean
weight of the non-aboriginal population is higher than the aboriginal population (70.87 ±
26.31 kg vs. 65.77 ± 30.45 kg), yet the non-aboriginal population has a lower mean body
mass index (25.92 ± 6.33 kg/m^ vs. 27.25 ± 8.02 kg/m^).
Since both weight and body mass index values were available for analysis in this
study, the relationships between both of these factors and diabetic status were explored.
There was a general trend observed in both the aboriginal and non-aboriginal populations; as
the weight increased so did the prevalence of diabetes. ANOVA analysis of the weight factor
in relation to diabetic status revealed that, overall, weight was significantly related to diabetic
status in both the aboriginal and non-aboriginal populations. The univariate logistic
regression model indicated that all of the weight categories were significantly related to an
increased risk of diabetes (Table 7). However, the multivariate logistic regression model
indicated that having a weight of between 60 kg and 69 kg does not significantly increase
one’s risk of developing diabetes (p = 0.413). While there is a significant risk associated
with weights of 70 kg and greater, these risks are lower in the multivariate model than the
univariate model suggesting that after controlling for the potentially confounding effects of
gender, age, and aboriginal status, weight makes less of a contribution to the overall risk of
developing diabetes.

52

As previously mentioned, body mass index is the measure of obesity most commonly
reported in the literature. Therefore, it is on this basis that we are more easily able to compare
the results of this study with other studies looking at the risk factors associated with diabetes.
According to a report by Health Canada (2002), approximately 38.0% of women and 59.7%
of men, aged 20 to 59 years, reported being overweight in Canada in 1998/99. Of these
cohorts of overweight individuals, 76.2% of the women and 72.9% of the men self-reported
having diabetes. Choi and Shi (2001), in their study of the risk factors associated with
diabetes, also found that the prevalence of diabetes increased with body mass index and that
the odds ratios for body mass index, decreased from the univariate model to the multivariate.
This is a trend that was seen in the Bella Coola Valley study in relation to weight, again
supporting the notion that while obesity is an important contributor to the overall risk of
developing diabetes, its contribution is somewhat diminished when other confounders are
controlled for. Harris et al. (2003), in their study of the Ojibwa-Cree in northwest Ontario,
found that body mass index was significantly associated with diabetes in both men and
women in the 18-49 years age group, yet not in the greater than or equal to 50 years age
group. They also reported that generally for a 5 kg/m^ increase in body mass index the odds
ratio was 2.39 (95% Cl, 1.32 - 4.35). This is lower than the odds ratios determined in the
study of the Bella Coola Valley population for body mass index. In this study, using a body
mass index of less than 25 kg/m^ as the reference, the odds ratios ranged from 7.0 (95% Cl,
3.1 - 16.2) in the 25 - 29.9 kg/m^ category to 56.1 (95% Cl, 23.2 - 135.4) in the greater than
or equal to 40 kg/m^ category. Another study of the Ojibwa-Cree found that the odds ratio
for a body mass index greater than or equal to 26 kg/m^, when compared to a body mass
index of less than 26 kg/m^, was 4.51 (95% Cl, 1.46 - 6.74) (Young, Sevenhuysen, Ling, &

53

Moffatt, 1990). These results are more similar to the results seen in the Bella Coola Valley
study than the previously mentioned study by Harris et al., yet is still not as high as the odds
ratios found in the Bella Coola Valley.
Type 2 diabetes is a chronic condition that affects over a million adults in Canada
(Health Canada, 2002). Given the chronic nature of diabetes and the high incidence of
subsequent complications and comorbidities seen in people with diabetes, this disease is
associated with tremendous cost to both individuals and society as a whole. Many aboriginal
groups across Canada are experiencing higher prevalence rates and greater severity of
disease, while for others, type 2 diabetes is less frequent (Young, Szathmary et al., 1990).
The variation seen in the prevalence rates of diabetes across different aboriginal groups
indicates the need to study diabetes within the context of individual aboriginal groups. As
shown by the discrepancy between what has been reported as the overall prevalence rate of
diabetes in British Columbian aboriginal groups versus what was found in this study
regarding the prevalence in the Bella Coola Valley, broad surveys of populations are not an
accurate depiction o f the problem of diabetes for individual aboriginal groups. The reported
prevalence rates of diabetes in British Columbia for aboriginal people are very low by
comparison to other aboriginal groups across Canada, and even lower than the prevalence
rate reported for the general population of Canada (Johnson et al., 2002; Martin &
Yidegiligne, 1998; Young, Szathmary et al., 1990). Based on this information, one could
draw the conclusion that diabetes is not a significant issue for the aboriginal people of British
Columbia, and therefore, is not a priority to be addressed at the community level or within
the health care system. However, the study of the Bella Coola Valley shows that diabetes is

54

indeed a significant health problem facing this population, especially for the aboriginal
people, and that it must be addressed promptly.

55

Chapter Six
Implications of this Study

The results of this study indicate that diabetes is a significant health problem that
needs to be addressed within the Bella Coola Valley community. The advantage in dealing
with type 2 diabetes is that many of the risk factors associated with the disease are
modifiable. While a person’s age or aboriginal status cannot be changed, a person’s weight
and body mass index can be. For this reason most prevention strategies for type 2 diabetes
centre around lifestyle modifications involving diet and exercise. In addition to focusing on
individual behaviour changes, it is important that any prevention strategy also addresses the
environment in which these people live and work, to make it supportive of healthy lifestyle
choices. When working with aboriginal people, it is equally important to ensure that any
intervention strategy is culturally sensitive to the traditions and beliefs of the community.
One way to ensure that this occurs is to include representatives from the aboriginal group
throughout the planning and implementation stages, thereby empowering them to be active
participants in the intervention strategy.
This approach has been proven suecessfiil, within the context of diabetes, in such
projects as the Kahnawake Sehools Diabetes Prevention Project and the Sioux Lookout
Diabetes Program. The Kahnawake Schools Diabetes Prevention Project is working to
improve the healthy eating habits and increase the physical activity of the residents of
Kahnawake, Quebec. There are two components to this project: the school component that
targets elementary school children, and the community component that aims to foster
community involvement and support. The school component, which consists of a “culturally
relevant Health Edueation Program”, has been the catalyst for other school-wide initiatives

56

including a Nutrition Policy which ensures ehildren are exposed only to nutritious foods
while in school (Macaulay et ah, 1998). The community has been actively involved in this
project through the development of the Community Advisory Board, which has been a
driving force in the planning, development and implementation of the overall program. It is
thought by the authors that the strong community involvement and holistic approach are
some of the major key components to the program’s success. The Sioux Lookout Diabetes
Program addresses both diabetes prevention and management through a variety of
approaches including diabetes education, foot care services, and nutrition advice that
includes a grocery store tour and budget food guide. Innovative programs such as a summer
camp for youth with type 2 diabetes have been used to address the specific needs of the
population within the context of their own culture (Morrison & Dooley, 1996).
This study can be used in support for the need to develop community programs to
address diabetes in the Bella Coola Valley. The community could learn from such programs
as those previously mentioned and work to establish their own community-based diabetes
program. The Nuxalk Band Council has already taken an active interest in the health of their
people and has shown their willingness to work with health care providers and researchers to
improve the health of their community.

57

Chapter Seven
Future Research

This study has added to the limited hody of literature on the prevalence of diabetes
among the aboriginal people of British Columbia. It has also illustrated the need for
individual specific study of different aboriginal groups, given the stark variation in
prevalence rates observed. For the purposes of community level planning of programs and
health care resources, province-wide surveys do not provide an accurate picture of the
problem of diabetes among aboriginal communities.
Further research is required to gain a more comprehensive understanding of diabetes
in the Bella Coola Valley and the modifiable risk factors most pertinent to this population
such as obesity and physical inactivity. Another risk factor that should be explored is the
role of family history in the development of diabetes in this community. The high prevalence
of diabetes among aboriginal women aged 25 to 39 years observed in this study indicates that
the potential link between the development of diabetes and maternal diabetes needs to be
explored further in the Bella Coola Valley.

58

References

American Diabetes Association. (2002). Evidence-based nutrition principles and
recommendations for the treatment and prevention of diabetes and related
complications. D iabetes Care, 25(Supplement 1), S50-S60.
American Diabetes Association. (2003). D iabetes Dictionary. Retrieved December 2, 2003,
2003, from www.diabetes.org/info/dictionarv.jsp?pageID=l&exitDictionarvTo=
Arslanian, S. (2002). Type 2 diabetes in children: Clinical aspects and risk factors. Hormone
Research, J/(Supplement 1), 19-28.
Barker, D. J. P. (1999). The fetal origins of type 2 diabetes mellitus. Annals o f Internal
Medicine, 750(4 (Part 1)), 322-324.
BC Stats. (2002). Regional D istrict 45 - Central Coast Statistical Profile. Retrieved January
16, 2004, 2004, from www.bcstats.gov.bc.ca/data/sep/rd/Rd__45.pdf
Bella Coola General Hospital, (n.d.). Welcome to the Bella Coola General Hospital.
Retrieved January 16, 2004, 2004, from http://bellacoola hospital.tripod.com
Bella Coola Valley Tourism. (2003). Bella Coola. Retrieved January 16, 2004, 2004, from
www.britishcolumbia.com/regions/towns/?townID=3587
Benyshek, D. C., Martin, J. P., & Johnston, C. S. (2001). A reconsideration of the origins of
the type 2 diabetes epidemic among Native Americans and the implications for
intervention policy. M edical Anthropology, 20, 25-64.
Brassard, P., Robinson, E., & Lavallee, C. (1993). Prevalence of diabetes mellitus among the
James Bay Cree of northern Quebec. Canadian M edical Association Journal, 149(3),
303-306.
Burrows, N. R., & Geiss, L. S. (2000). Prevalence of diabetes in among Native Americans
and Alaska Natives, 1990-1997: An increasing burden.
Care, 23(12), 17861790.
Canadian Diabetes Association. (2000, June 2000). D iabetes. Retrieved December 2, 2003,
2003, from www.diabetes.ca/files/DiabetesFactSheet.pdf
Central Coast Economic Development Commission. (1990). Bella Coola, British Columbia,
Canada. Bella Coola: The Job Development Project.
Choi, B. C. K., & Shi, F. (2001). Risk factors for diabetes mellitus by age and sex: results of
the National Population Health Survey. Diabetologia, 44, 1221-1231.

59

Dabelea, D., Knowler, W. C., & Pettitt, D. J. (2000). Effect of diabetes in pregnancy on
offspring: Follow-up research in the Pima Indians. Journal o f M aternal-Fetal
Medicine, P, 83-88.
Delisle, H. P., & Ekoe, J. (1993). Prevalence of non-insulin-dependent diabetes mellitus and
impaired glucose tolerance in two Algonquin communities in Quebec. Canadian
M edical Association Journal, 148{\), 41-46.
Diabetes Working Group. (2002). Improving Chronic D isease Management: A Compelling
Business Case fo r Diabetes. Victoria: Government of British Columbia.
Dyck, R. F., Klomp, H., & Tan, L. (2001). From "thrifty genotype" to "hefty fetal
phenotype": The relationship between high birthweight and diabetes in Saskatchewan
Registered Indians. Canadian Journal o f Public Health, 92(5), 340-344.
Evers, S., McCracken, E., Antone, I., & Deagle, G. (1987). The prevalence of diabetes in
Indians and Caucasians living in southwestern Ontario. Canadian Journal o f Public
7g, 240-243.
Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. (2002). Report
of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus.
D iabetes Care, 25(Supplement 1), 85 - 820.
First Nations and Inuit Elealth Branch. (2003). A Statistical Profile on the Health o f First
Nations in Canada. Ottawa: Health Canada.
Fletcher, B., Gulanick, M., & Lamendola, C. (2002). Risk factors for type 2 diabetes
meWiius. Journal o f Cardiovascular Nursing, 16{2), 17-23.
Gobrial, M., Mekael, H., Anderson, N., Ayers, D., & Thommasen, H. V. (2002). Diabetic
blood sugar control: An urban/rural comparison. British Columbia M edical Journal,
4^/(10), 537-543.
Hales, C. N., Barker, D. J. P., Clark, P. M. 8., Cox, L. J., Fall, C., Osmond, C., et al. (1991).
Fetal and infant growth and impaired glucose tolerance at age 64. British M edical
Journal, 303, 1019-1022.
Harris, 8. B., Gittlesohn, J., Hanley, A., Bamie, A., Wolever, T. M. 8., Gao, J., et al. (2003).
The prevalence of NIDDM and associated risk factors in native Canadians. D iabetes
Care, 20(1), 185-187.
Health Canada. (2000). D iabetes Among Aboriginal People in Canada: The Evidence.
Ottawa: Health Canada.
Health Canada. (2002). D iabetes in Canada: Second Edition. Ottawa: Center for Chronic
Disease Prevention and Control, Population and Public Health Branch.

60

Hennekens, C. H., & Buring, J. E. (1987). Epidemiology in medicine. Philadelphia:
Lippincott Williams & Wilkins.
Hernandez, C., Antone, I., & Cornelius, I. (1999). A grounded theory study of the experience
of type 2 diabetes mellitus in First Nations adults in Canada. Journal o f Trans cultural
Nursing, 10(f), 220-228.
Johnson, S., Martin, D., & Sarin, C. (2002). Diabetes mellitus in the First Nations population
of British Columbia, Canada: Part 3. Prevalence of diagnosed eases. International
Journal o f Circumpolar Health, 61, 260-264.
Jorgensen, M. E., Bjeregaard, P., Boreh-Johnsen, K., Backer, V., Becker, U., Jorgensen, T.,
et al. (2002). Diabetes and impaired glucose tolerance among the Inuit population of
Greenland. D iabetes Care, 25(10), 1766-1771.
Kelsey, J. L., Whittemore, A. S., Evans, A. S., & Thompson, W. D. (1996). M ethods in
observational epidemiology (2nd ed. Vol. 26). New York: Oxford University Press.
Levin, S., Mayer-Davis, E. J., Ainsworth, B. E., Addy, C. L., & Wheeler, F. C. (2001).
Behavioral and demographic correlates of diabetes, hypertension and overweight
among the Catawba Indian Nation. Ethnicity & Disease, 11, 241-250.
Lieberman, L. S. (2003). Dietary, evolutionary, and modernizing influences on the
prevalence of type 2 diabetes. Annual Review o f Nutrition, 23, 345-377.
Maberley, D. A. L., King, W., & Cruess, A. F. (2000). The prevalence of diabetes in the Cree
of Western James Bay. Chronic D iseases in Canada, 27(3), 128-133.
Macaulay, A. C., Montour, L. T., & Adelson, N. (1988). Prevalence of diabetic and
atherosclerotic complications among Mohawk Indians of Kahnawake, PQ. Canadian
M edical Association Journal, 139, 221-224.
Macaulay, A. C., Paradis, G., Potvin, L., Cross, E. J., Saad-Haddad, C., McComber, A. M., et
al. (1998). Primary prevention of type 2 diabetes in First Nations: Experiences of the
Kahnawake Sehools Diabetes Prevention Project. Canadian Journal o f D iabetes
Care, 22(3), 44-49.
Martin, J. D., & Yidegiligne, H. M. (1998). Diabetes mellitus in the First Nations population
of British Columbia, Canada. International Journal o f Circumpolar Health, 57(Suppl
1),335-339.
Meltzer, S., Leiter, L., Daneman, D., Gerstein, H. C., Lau, D., Ludwig, S., et al. (1998). 1998
clinical practice guidelines for the management of diabetes in Canada. Canadian
M edical Association Journal, 759(Suppl. 8), S1-S29.

61

Morrison, N., & Dooley, J. (1996). The Sioux Lookout Diabetes Program: Diabetes
prevention and management in northwestern Ontario. Circumpolar Health, 96, 364369.
Motala, A. A., Pirie, F. J., Gouws, E., Amod, A., & Omar, M. A. K. (2003). High ineidenee
of Type 2 diabetes mellitus in South African Indians: a 10-year follow-up study.
D iabetic Medicine, 20(1), 23-30.
National Center for Chronic Disease Prevention and Health Promotion. (2003, February 26,
2003). Prevalence o f diagnosed diabetes p e r 100 population, by age. United States,
1980-2000. Retrieved November 13, 2003, 2003, from
W W W , cdc. gov/ diabetes/statistics/prev/national/tableS .htm
Neel, J. V. (1962). Diabetes mellitus: A "thrifty" genotype rendered detrimental by
"progress". American Journal o f Human Genetics, 14, 353-362.
Pioro, M. P., Dyck, R. F., & Gillis, D. C. (1996). Diabetes prevalence rates among First
Nations adults on Saskatchewan reserves in 1990: comparison by tribal grouping,
geography and with Non-First Nation people. Canadian Journal o f Public Health,
g7(5), 325-328.
Puska, P., Nisbida, C., & Porter, D. (2003). Obesity and Overweight. Retrieved December 2,
2003, 2003, from www.wbo.int/bpr/NPH/docs/gs_obesitv.pdf
Ricb-Edwards, J. W., Colditz, G. A., Stampter, M. J., Willett, W. C., Gillman, M. W.,
Hennekens, C. H., et al. (1999). Birthweight and the risk for type 2 diabetes mellitus
in adult women. Annals o f Internal Medicine, 130, 278-284.
Rodrigues, S., Robinson, E. J., Kramer, M. S., & Gray-Donald, K. (2000). High rates of
infant macrosomia: A comparison of a Canadian native and non-native population.
Journal o f Nutrition, 130, 806-812.
Schulz, L. O. (1999). Traditional environment protects against diabetes in Pima Indians.
Healthy Weight Journal, 13{5), 68-70.
Sbeard, N. F. (2003). Moderate changes in weight and physical activity can prevent or delay
the development of type 2 diabetes mellitus in susceptible individuals. Nutrition
Reviews, 61(2), 76-79.
Statistics Canada. (2001, December 12, 2003). 2001 Community Profiles. Retrieved January
12, 2004, 2004, from
WWW12. statcan.ca/englisb/profil 01/S earcb/PlaceS earcbF orm l.cfrn
Thommasen, H. V., Loewen, T., & Mclnnes, K. (1995). Medicinal plants of the Nuxalk
natives of Bella Coola. British Columbia M edical Journal, 37(6), 403-408.

62

Thommasen, H. V., Newberry, P., & Watt, W. D. (1999). Medieal history of the central coast
of British Columbia. British Columbia M edical Journal, 41(9), 464-470.
Vaaler, S. (2000). Optimal glycémie control in type 2 diabetic patients: Does including
insulin treatment mean a better outcome? D iabetes Care, 25(Supplement 2), B30B34.
Vancouver Coastal Health Authority. (2003). Bella Coola General H ospital Program
Description. Retrieved January 16, 2004, 2004, from
https ://www.vcha.net/Progi~ams/prog_loc.asp?prog_loc=711 &prog= 1372&loc= 1835
Wendorf, M. (1992). Archaeology and the "thrifty" non insulin dependent diabetes mellitus
(NIDDM) genotype. Advances in Peritoneal Dialysis. Conference on Peritoneal
Dialysis, 8, 201-207.
Young, T. K. (1994). The Health o f N ative Americans: Towards a Biocultural Epidemiology.
New York: Oxford University Press.
Young, T. K., & Harris, S. B. (1994). Risk of clinical diabetes in a Northern Native Canadian
cohort. A rctic M edical Research, 53, 64-70.
Young, T. K., Martens, P. J., Taback, S. P., Sellers, E. A. C., Dean, H. J., Cheang, M., et al.
(2002). Type 2 diabetes mellitus in Children: Prenatal and early infancy risk factors
among Native Canadians. Archive o f Pediatric and Adolescent Medicine, 156, 651655.
Young, T. K., McIntyre, L. L., Dooley, J., & Rodriguez, J. (1985). Epidemiologic features of
diabetes mellitus among Indians in northwestern Ontario and northeastern Manitoba.
Canadian M edical Association Journal, 132, 793-797.
Young, T. K., Moffatt, M. E. K., & Ling, N. (1988). Risk factors for diabetes among northern
Indians. Arctic M edical Research, 4 /(Supplement 1), 445-448.
Young, T. K., Reading, J., Elias, B., & O'Neil, J. D. (2000). Type 2 diabetes mellitus in
Canada's First Nations: status of an epidemic in progress. Canadian M edical
Association Journal, 163(5), 561-566.
Young, T. K., Sevenhuysen, G. P., Ling, N., & Moffatt, M. E. K. (1990). Determinants of
plasma glucose level and diabetic status in a northern Canadian Indian population.
Canadian M edical Association Journal, 742(8), 821-830.
Young, T. K., Szathmary, E. J. E., Evers, S., & Wheatley, B. (1990). Geographical
distribution of diabetes among the native population of Canada: A national survey.
Social Science & Medicine, 31(2), 129-139.

63

Zimmet, P. (1982). Type 2 (non-insulin-dependent) diabetes: An epidemiologieal overview.
Diabetologia, 22, 399-411.

64

Appendix I

Ethics Approval Certificate

65

Appendix II
Direct Age Adjustment of Prevalenee Rates

Age
1 8 -2 4 .9
25 - 39.9
40 - 44.9
45 - 64.9
65+
Total

Aboriginal
Population
# with diabetes
0
140
254
7 (2.75)*
72
5 (6.94)
195
38 (19.49)
58
20 (34.48)
719
70

Non-aboriginal
# with diabetes
Population
0
92
247
0
113
3 (2.65)
394
30(7.61)
160
24(15)
1006
57

Numbers in parentheses represent the prevalence rate per 100

Calculation of prevalence rate per 100:

254

X100 = 2.75

Age
1 8 -2 4 .9
25 - 39.9
40 - 44.9
45 - 64.9
65+
Total

Total Population*
232
501
185
589
218
1725

# cases expected
Aboriginal
0
13.78
12.84
114.80
75.17
216.59

# cases expected
Non-aboriginal
0
0
4.90
44.82
32.7
82.42

Total population calculated by adding the aboriginal and non-aboriginal populations for each age group

Calculation of expected # of cases:
2.75x501
= 13.78
100
Calculation of age-adjusted prevalence rates:
Aboriginal

216.59
xlOO = 12.55
1725

Non-aboriginal

82.42
X100 = 4.78
1725

67