The correlates of current smoking among adult Metis: evidence from the Aboriginal Peoples Survey and Metis supplement.
Ryan, Christopher J. ; Cooke, Martin J. ; Leatherdale, Scott T. 等
Aboriginal Canadians suffer from poorer health than non-Aboriginal
Canadians. (1) Among the most notable medical conditions that
characterize this diminished health status are cardiovascular and
pulmonary diseases. (1, 2) It is well established that smoking, a
behaviour more prevalent among Aboriginal Canadians compared to their
non-Aboriginal counterparts, (3) is an important contributory factor to
these diseases. (4)
Prior research has suggested that cultural identity and practices
are important to the health of Aboriginal peoples. (5) Retention of
Aboriginal languages, participation in traditional activities,
environmental and cultural connections, and spirituality have been
identified as components of an Aboriginal-specific health framework.
(5-7) One could therefore hypothesize that these determinants of health
might be connected to health behaviours among Aboriginal peoples,
including the Metis.
The Metis, a distinct and constitutionally recognized Aboriginal
Canadian group with a population of close to 400,000 people, (8) like
other Aboriginal Canadians, are at higher risk of suffering from chronic
health conditions, such as respiratory and cardiovascular disease, than
are non-Aboriginal Canadians. (8) Recent data revealed that 30% of Metis
over the age of 12 reported smoking cigarettes daily, approximately two
times greater than the daily smoking rate of non-Aboriginal Canadians.
(3) However, there is a paucity of research that has examined the
correlates of smoking among Metis, particularly with a focus on
culturally-specific factors.
A significant body of literature exists on the correlates of
smoking among the general population. Those with low income and
education are more likely to smoke. (9) In addition, low physical
activity levels, heavy drinking, and low body mass index are
health-related variables correlated with smoking among the general
population. (10-12) Prior research has also demonstrated that spiritual
well-being and religiosity might be protective against smoking among
adults in the general population. (13, 14)
Given the importance of cultural identity to health, (5) the
historical and cultural significance of tobacco among Aboriginal
peoples, (15, 16) and the potential negative association between
spirituality and smoking, (13, 14, 17) an examination of how
culturally-specific factors are correlated with smoking among Metis is
warranted to guide the development of public health interventions to
decrease rates of smoking among this population. Guided by an
Aboriginal-specific determinants of health framework, (5-7) and prior
evidence on the correlates of smoking among the general population,
(10-14) this study used data from the 2006 Aboriginal Peoples Survey
(APS) and Metis Supplement to investigate factors associated with
current smoking among Metis 18 years of age and older. *
METHODS
Data and sample
The APS, a postcensal survey, was administered to a sample of
Aboriginal Canadians, including Metis, First Nations people living
off-reserve and Inuit. (18) Despite the availability of the 2012 APS,
the 2006 version is the most recent to include, through the Metis
Supplement, measures of spirituality, membership in a Metis
organization, and attendance at Metis cultural events. A total of 8,830
Metis aged 15 years and older responded to the 2006 APS and Metis
Supplement, including 8,160 aged 18 and older. Once respondents with
missing responses and 110 women who reported being pregnant and
therefore missing BMI measures, were excluded, 6,610 Metis were included
in the sample. Less than 1% of the sample had missing responses on the
smoking measure. For the independent variables, the combination of
"don't know", "not stated" and
"refused" responses ranged from 0% to 4.1%. We determined that
missing cases could be safely treated as missing at random after
examining the relationships between non-response and the outcome
variable.
Variables
The outcome variable was measured by a binary variable, current
smoking, which includes daily and occasional smoking. (19) The APS
assessed current smoking via the question "at the present time do
you smoke cigarettes daily, occasionally or not at all?"
Several culturally-specific factors were considered based on the
components of an Aboriginal-specific determinants of health framework,
(5-7) including level of spirituality, knowledge of an Aboriginal
language, membership in a Metis organization, and last time attending a
Metis cultural event or a pilgrimage, or having seen Metis artists
perform. Level of spirituality was coded into four categories:
"very religious or spiritual", "moderately religious or
spiritual", "not very religious or spiritual", and
"not at all religious or spiritual". We measured knowledge of
an Aboriginal language by combining two APS questions: "Do you
speak an Aboriginal language?" and "Is any Aboriginal
language, such as Michif, Cree, Saulteax, or Dene, ever spoken at
home?" As a result, knowledge of an Aboriginal language was coded
into three categories: "speaks an Aboriginal language",
"does not speak an Aboriginal language but an Aboriginal language
is spoken at home", and "neither speaks an Aboriginal language
nor lives in a home where an Aboriginal language is spoken".
Membership in a Metis organization was assessed by a binary variable
(yes/no) based on the APS response categories. Finally, last time having
attended a Metis cultural event was categorized into four levels based
on the sample distribution: "less than 1 year ago", "from
1 to 5 years ago","5 or more years ago", and
"never".
Demographic, geographic, socio-economic and health-related
covariates were selected based on prior research showing significant
associations between these factors and smoking among the general
population. (9-11, 20, 21) Age was grouped into five categories (18-19
years, 20-34 years, 35-49 years, 50-64 years, and 65 years and older),
similar to previous research that has examined smoking among the general
Canadian population. (20) Urban/rural geography was coded into four
categories: "Census Metropolitan Areas (CMAs)", ([dagger])
"Census Agglomerations (CAs)", ([double dagger]) "Rural
with moderate to high Metropolitan Influence (MIZ)", ([section]) or
"Rural with no to weak MIZ". Regional geographic
classifications included in the analyses were "Atlantic",
"Quebec", "Ontario", "Prairies",
"British Columbia" and "Territories". Annual
household income was divided into quartiles (less than $35,000,
$35,000-$60,899, $60,900-$95,899, and greater than $95,899). Educational
attainment was grouped into three categories: "less than high
school diploma", "high school diploma or high school
equivalency" or "some post-secondary or more".
The health-related variables considered in this study were
self-perceived health, heavy drinking, body mass index (BMI) and
leisure-time physical activity (LTPA). Self-perceived health was
categorized as "excellent", "very good",
"good" and "fair or poor". Heavy drinking was
dichotomized as "consumes 5 or more drinks once per month or
more" or "does not consume 5 or more drinks once per month or
more", according to the definition of heavy drinking that was
employed by Statistics Canada for men and women from 2001 to 2012. (23)
For women, drinking 4 or more drinks once per month or more is
considered heavy drinking, by recent guidelines accepted by Statistics
Canada. (23) However, the 2006 APS question relating to alcohol
consumption prevented us from using the new definition of heavy drinking
for women. (22) For BMI, standard cutoffs for adults aged 18 and older
were used to create "underweight or normal weight",
"overweight" and "obese" categories. (24) LTPA was
classified as participating in 3 or more hours (180 minutes) of physical
activity per week or less than 3 hours per week, a cutoff close to,
although more stringent than, the 150 minutes of moderate to vigorous
physical activity per week recommended by the Canadian Society of
Exercise Physiology. (25)
Analyses
Chi square tests of independence were used to examine the sample
distribution and to test for bivariate associations. Multivariate
analyses were conducted using a sequential logistic regression method.
The first model included only culturally-specific factors. Subsequent
models were estimated with the addition of various blocks of variables
to test for associations between culturally-specific factors and current
smoking independent of potential confounders. Demographic and geographic
variables were added to the second model. Socio-economic and
health-related variables were added to the third and fourth models
respectively. We also estimated a model with age and gender interaction
terms. All models were estimated using the bootstrap weights that were
specially designed for the APS to account for the survey's complex
sampling design.
RESULTS
Overall, 39.9% of adult Metis respondents in the sample reported
being current smokers. Table 1 presents the sample characteristics and
bivariate associations between the independent variables and current
smoking. As predicted, bivariate associations show that those who
reported high levels of spirituality were less likely to smoke.
Respondents who were members of a Metis cultural organization were also
somewhat less likely to smoke compared to non-members. On the other
hand, respondents who spoke an Aboriginal language or lived in a
household where one was spoken were more likely to smoke than those who
neither spoke nor lived in a house where an Aboriginal language was
spoken. In addition, those who had more recently attended a Metis
cultural event were more likely to smoke (Table 1).
In general, the directions of the bivariate associations between
current smoking and health-related variables were as expected. Those who
were more physically active, had higher ratings of self-perceived
health, and higher BMI were less likely to smoke, whereas those who
drank heavily were more likely to smoke (Table 1).
The multivariate results presented below relate to the full model
(Table 2, model IV), which includes culturally-specific, demographic,
geographic, socio-economic and health-related variables.
Consistent with the bivariate results, respondents who reported
speaking an Aboriginal language were more likely to smoke (OR = 1.80,
95% CI 1.45-2.24) compared to those who did not speak an Aboriginal
language and did not live in a home where an Aboriginal language was
spoken. In addition, those who reported not speaking an Aboriginal
language but living in a home where an Aboriginal language was spoken
were more likely to smoke (OR = 1.46, 95% CI 1.11-1.95) compared to
those who neither spoke an Aboriginal language nor lived in a home where
an Aboriginal language was spoken.
As hypothesized, Metis adults who reported higher levels of
spirituality were less likely to smoke. In particular, adult Metis who
reported being "very religious or spiritual" or
"moderately religious or spiritual" were less likely to smoke
(OR = 0.74, 95% CI 0.58-0.93; OR = 0.78, 95% CI 0.62-0.96 respectively)
compared to those who were "not at all religious or
spiritual". Attending Metis cultural events and being a member of a
Metis organization were not significantly correlated with current
smoking.
Women were significantly more likely to smoke (OR = 1.23, 95% CI
1.07-1.42) than men. Respondents aged 18-19 were less likely to smoke
(OR = 0.55, 95% CI 0.40-0.74) than those aged 20-34. Similarly,
respondents aged 50-64 and over 65 were significantly less likely to
smoke (OR = 0.67, 95% CI 0.54-0.82; OR = 0.24, 95% CI 0.17-0.34
respectively) compared to adult Metis aged 20-34.
Although urban/rural geography was not significantly associated
with smoking, Metis living in the Prairies were more likely to smoke
compared to those living in Ontario (OR = 1.23, 95% CI 1.02-1.49).
As expected, annual household income and educational attainment
were negatively associated with current smoking. Metis with a household
income in the first, second and third lowest quartiles were more likely
to smoke (OR = 2.32, 95% CI 1.90-2.85; OR = 1.84, 95% CI 1.53-2.22; OR =
1.25, 95% CI 1.04-1.51 respectively) relative to respondents with an
income in the highest quartile. Respondents with less than a high school
education were more likely to smoke (OR = 1.86, 95% CI 1.562.22) than
those who had attained some post-secondary education or more. Metis who
had completed high school were also more likely to smoke (OR = 1.42, 95%
CI 1.19-1.69) than those with at least some post-secondary education.
Adult Metis who reported excellent and very good self-perceived
health were less likely to smoke than those with poor health (OR = 0.41,
95% CI 0.32-0.52; OR = 0.70, 95% CI 0.57-0.87 respectively). In
addition, those who reported participating in 3 or more hours of LTPA
per week were less likely to smoke (OR = 0.78, 95% CI 0.69-0.89) than
respondents who reported less than 3 hours of LTPA per week. Respondents
who reported heavy drinking were more likely to smoke (OR = 1.84, 95% CI
1.55-2.18) than those who did not report drinking heavily. Compared to
normal weight respondents, those who were overweight or obese were less
likely to smoke (OR = 0.60, 95% CI 0.51-0.70; OR = 0.41, 95% CI
0.34-0.49 respectively).
In a separate model-building procedure, we estimated a model to
investigate how age and gender interact with culturally-specific
variables (results not shown). Although this model did not substantially
change the interpretation of the results of this study, a significant
interaction term was found between age and spirituality. While
spirituality was independently associated with smoking at all ages,
respondents aged 50-64 who reported being "not very spiritual"
were significantly more likely to smoke than those in other age groups
who reported the same level of spirituality (Figure 1). The full table
of this model is available from the authors on request.
DISCUSSION
This study examined the correlates of current smoking among Metis
18 years of age and older using data from the 2006 Aboriginal Peoples
Survey and Metis Supplement. To our knowledge, this is the first study
to examine the associations between culturally-specific factors and
smoking among this population. The associations between smoking and
various demographic, geographic, socio-economic and health-related
factors were also considered.
[FIGURE 1 OMITTED]
Two fifths of Metis 18 years of age and older were current smokers
in 2006, more than double the 19% smoking rate in 2006 found in the
general Canadian population aged 15 and older. (26) Similarly, research
using data combined from the 2007 to 2010 CCHS cycles found that 30% of
Metis over the age of 12 were daily smokers, compared to 15% of the
non-Aboriginal population. (3) The high rate of smoking among Metis
clearly demonstrates the need to develop a better understanding of what
predicts smoking among this population to guide effective public health
interventions.
A particularly interesting finding in the current study was that a
high level of spirituality or religiosity was negatively associated with
current smoking among the Metis respondents in this sample.
Specifically, adult Metis who reported being "very" or
"moderately" spiritual or religious were significantly less
likely to smoke than those who were "not at all" religious or
spiritual. Although these findings might be explained by social factors
not captured in the models, they are intriguing for a number of reasons.
First, spirituality has been identified as an important constituent of
an Aboriginal-specific health framework. (5-7) Second, among the general
population, prior research has demonstrated that smoking is less
prevalent among those who report higher spiritual well-being and more
religious activity. (13, 14) This is perhaps because those with higher
levels of religiousness or spirituality benefit from stronger social
support, positive coping mechanisms against stress, and discouragement
to smoke from religious leaders. (14) However, further research is
needed to determine how these mechanisms might be related to current
smoking among adult Metis.
Contrary to the significant negative association between
spirituality and smoking, we found that adult Metis who spoke an
Aboriginal language or lived in a household where an Aboriginal language
was spoken were more likely to be current smokers, independent of the
covariates included in the analyses. However, these results do not
suggest that speaking an Aboriginal language or living in a household
where one is spoken are, in themselves, risk factors for smoking.
Language is a social phenomenon and a strong component of Aboriginal
identity. (27) In this case, the relationship between Aboriginal
language and current smoking might be explained by social factors not
captured in the models. Among possible directions for future research,
qualitative studies using focus group methodology could help to
contextualize the relationships among spirituality, Aboriginal language
and current smoking among Metis.
With regard to socio-economic and health-related covariates, many
factors associated with smoking among Metis are similar to what has been
observed among the general population, including low income, low
education, (9) low levels of physical activity, (10) and heavy alcohol
consumption. (11) Although smoking cessation interventions should target
all Metis who smoke, these findings emphasize that interventions should
particularly target those with low income and education, and those who
engage in other health behaviours that are not health-promoting. In
addition, our results demonstrated that age and gender were correlates
of current smoking, suggesting that age- and gender-specific
interventions should be considered among Metis.
There are a number of limitations associated with the APS data. One
is that the data are observational and cross-sectional, making it
impossible to infer causality. In addition, there is an inflated
possibility of type 1 error, since many comparisons were made in this
study. Another limitation is that smoking is self-reported, creating the
possibility for reporting bias. However, recent literature that used
data from the 2007 to 2009 Canadian Health Measures Survey has suggested
that self-reporting is a valid tool for measuring smoking status. (28)
There is also the possibility that the sample of Metis included in the
APS is somewhat biased. Since the APS is a postcensal survey, those who
did not respond to the Census would not have been included in the sample
for the APS. This is a concern since the population least likely to
respond to the Census is one that is young, highly mobile and of lower
income, characteristics more common among Metis than among the general
population. (29) Nonetheless, the results of this study provide strong
evidence that interventions aimed at reducing the prevalence of smoking
among Metis should consider culturally-specific factors, such as
spirituality and Aboriginal language, as well as demographic,
socio-economic and health-related factors.
REFERENCES
(1.) Tjepkema M, Wilkins R, Senecal S, Guimond E, Penney C.
Mortality of Metis and registered Indian adults in Canada: An 11-year
follow-up study. Health Rep 2009; 20(4):31-51. PMID: 20108604.
(2.) Findlay LC. Physical activity among First Nations people off
reserve, Metis and Inuit. Health Rep 2011; 22(1):47-54. PMID: 21510589.
(3.) Gionet L, Roshanafshar G. Select Health Indicators of First
Nations People Living Off Reserve, Metis and Inuit. 2013. Available at:
http://www.statcan.gc.ca/pub/ 82-624-x/2013001/article/11763-eng.htm
(Accessed February 20, 2015).
(4.) Centers for Disease Control and Prevention (US), National
Center for Chronic Disease Prevention and Health Promotion (US), Office
on Smoking and Health (US). How Tobacco Smoke Causes Disease: The
Biology and Behavioral Basis for Smoking-Attributable Disease: A Report
of the Surgeon General. Atlanta, GA: Centers for Disease Control and
Prevention, 2010.
(5.) King M, Smith A, Gracey M. Indigenous health part 2: The
underlying causes of the health gap. Lancet 2009; 374(9683):76-85. PMID:
19577696. doi: 10.1016/ S0140-6736(09)60827-8.
(6.) Richmond CA, Ross NA. The determinants of First Nation and
Inuit health: A critical population health approach. Health Place
2009;15(2):403-11. PMID: 18760954. doi:
10.1016/j.healthplace.2008.07.004.
(7.) Wilson K, Rosenberg MW. Exploring the determinants of health
for First Nations peoples in Canada: Can existing frameworks accommodate
traditional activities? Soc Sci Med 2002; 55(11):2017-31. PMID:
12406468. doi: 10.1016/S0277-9536(01)00342-2.
(8.) Janz T, Seto J, Turner A. Aboriginal Peoples Survey, 2006: An
Overview of the Health of the Metis Population. Ottawa, ON: Statistics
Canada, 2009.
(9.) Reid JL, Hammond D, Driezen P. Socio-economic status and
smoking in Canada, 1999-2006: Has there been any progress on disparities
in tobacco use? Can J Public Health 2010; 101(1):73-78. PMID: 20364543.
(10.) deRuiter WK, Cairney J, Leatherdale ST, Faulkner GE. A
longitudinal examination of the interrelationship of multiple health
behaviors. Am J Prev Med 2014; 47(3):283-89. PMID: 25145617. doi:
10.1016/j.amepre.2014.04.019.
(11.) Paul LA, Grubaugh AL, Frueh BC, Ellis C, Egede LE.
Associations between binge and heavy drinking and health behaviors in a
nationally representative sample. Addict Behav 2011;36(12):1240-45.
PMID: 21868171. doi: 10.1016/j.addbeh.2011.07.034.
(12.) Plurphanswat N, Rodu B. The association of smoking and
demographic characteristics on body mass index and obesity among adults
in the U.S., 1999-2012. BMC Obesity 2014; 1(18). doi:
10.1186/s40608-014-0018-0.
(13.) Koenig HG, George LK, Cohen HJ, Hays JC, Larson DB, Blazer
DG. The relationship between religious activities and cigarette smoking
in older adults. J Gerontol A Biol Sci Med Sci 1998;53(6):M426-34. PMID:
9823746.
(14.) Garrusi B, Nakhaee N. Religion and smoking: A review of
recent literature. Int J Psychiatry Med 2012; 43(3):279-92. PMID:
22978085. doi: 10.2190/PM.43.3.g.
(15.) Health Canada. Tobacco: First Nations & Inuit Health.
2014. Available at: http://www.hc-sc.gc.ca/fniah-spnia/substan/tobac-tabac/index-eng.php (Accessed March 2, 2015).
(16.) Barkwell L, Dorion L, Hourie A, Dumont G. Metis Legacy. Vol.
2: Michif Culture, Heritage, and Folkways. Saskatoon, SK: Gabriel Dumont
Institute, 2006.
(17.) Franklin WJ. American African women and smoking: Spiritual
well-being makes a difference. J Christ Nurs 2011; 28(3):162-67. PMID:
21853718.
(18.) Statistics Canada. Aboriginal Peoples Survey, 2006: Concepts
and Methods Guide. 2009. Ottawa, ON. Available at:
http://www.statcan.gc.ca/pub/89-637-x/89637-x2008003-eng.htm (Accessed
January 18, 2015).
(19.) Statistics Canada. Health Profile. Statistics Canada
Catalogue No. 82-228-XWE. 2013. Ottawa, ON. Available at:
http://www12.statcan.gc.ca/health-sante/82228/index.cfm?Lang=E (Accessed
March 2, 2015).
(20.) Li FX, Robson PJ, Chen Y, Qiu Z, Lo Siou G, Bryant HE.
Prevalence, trend, and sociodemographic association of five modifiable
lifestyle risk factors for cancer in Alberta and Canada. Cancer Causes
Control 2009;20(3):395-407. PMID: 18998220. doi:
10.1007/s10552-008-9254-2.
(21.) Corsi DJ, Chow CK, Lear SA, Subramanian SV, Teo KK, Boyle MH.
Smoking in context: A multilevel analysis of 49,088 communities in
Canada. Am J Prev Med 2012; 43(6):601-10. PMID: 23159255. doi:
10.1016/j.amepre.2012.08.023.
(22.) Statistics Canada. Census Dictionary. Canada: Minister of
Industry. 2013. Available at:
https://www12.statcan.gc.ca/census-recensement/2011/ref/dict/
98-301-X2011001-eng.pdf (Accessed January 5, 2015).
(23.) Statistics Canada. Heavy Drinking. Statistics Canada
Catalogue No. 82-625-X. 2013. Ottawa, Ontario. Available at:
http://www.statcan.gc.ca/pub/82-625-x/ 2014001/article/14019-eng.htm.
(Accessed December 29, 2014).
(24.) Health Canada. Body Mass Index (BMI) Nomogram. 2012.
Available at: http://
www.hc-sc.gc.ca/fn-an/nutrition/weights-poids/guide-ld-adult/bmi_chart_
java-graph_imc_java-eng.php (Accessed January 27, 2015).
(25.) Canadian Society of Exercise Physiology (CSEP). Canadian
Physical Activity Guidelines and Canadian Sedentary Behaviour
Guidelines. 2012. Available at:
http://www.csep.ca/english/view.asp?x=949 (Accessed March 1, 2014).
(26.) Canadian tobacco use monitoring survey (CTUMS) 2012, 2013.
Available at: http://www.hc-sc.gc.ca/hc-ps/tobac-tabac/research-recherche/stat/ctumsesutc_2012-eng.php#tabb (Accessed January 28, 2015).
(27.) Frideres J. Continuity or disappearance: Aboriginal language
in Canada. In: Trovato F, Romaniuk A (Eds.), Aboriginal Populations:
Social, Demographic, and Epidemiological Perspectives. Edmonton, AB: The
University of Alberta Press, 2014; pp. 303-24.
(28.) Wong SL, Shields M, Leatherdale S, Malaison E, Hammond D.
Assessment of validity of self-reported smoking status. Health Rep 2012;
23(1):47-53. PMID: 22590805.
(29.) Statistics Canada. Aboriginal Statistics at a Glance. 2010.
Available at: http://www.
statcan.gc.ca/pub/89-645-x/89-645-x2010001-eng.htm (Accessed January 20,
2015).
Received: March 15, 2015
Accepted: May 16, 2015
Christopher J. Ryan, MSc, [1] Martin J. Cooke, PhD, [1, 2] Scott T.
Leatherdale, PhD, [1] Sharon I. Kirkpatrick, PhD, [1] Piotr Wilk, PhD
[3]
Author Affiliations
[1.] School of Public Health and Health Systems, University of
Waterloo, Waterloo, ON
[2.] Department of Sociology and Legal Studies, University of
Waterloo, Waterloo, ON
[3.] Department of Epidemiology and Biostatistics, University of
Western Ontario, London, ON
Correspondence: Christopher J. Ryan, School of Public Health and
Health Systems, University of Waterloo, Waterloo, ON N2L 3G1, Tel:
[telephone]519-573-3956, E-mail: cj2ryan@uwaterloo.ca
Conflict of Interest: None to declare.
* The data for this study were provided by Statistics Canada and
accessed at the South-Western Ontario Research Data Centre at the
University of Waterloo. The analyses and results are the authors'
alone.
([dagger]) A CMA is a geographic area that has a population of
100,000 people or more, with at least 50,000 people living in the core
area. (22)
([double dagger]) A CA is a geographic area that has a core
population of at least 10,000 people. (22)
([section]) A MIZ category of "strong",
"moderate", "weak" or "no" is assigned to
census subdivisions outside of CAs and CMAs based on the percentage of
the workforce that commute to a CA or CMA. (22)
Table 1. Sample characteristics by smoking status, Metis
18 years of age and older, Canada (2006 Aboriginal
Peoples Survey and Metis Supplement)
Current
smoker
No Yes p-
Characteristic N (%) (%) value
Last time having attended a
Metis cultural event
Less than 1 year ago 1830 58.5 41.5 0.0497
From 1 to 5 years ago 1490 59.1 40.9
5 or more years ago 1250 60.0 40.0
Never 2030 62.6 37.4
Speaks an Aboriginal language
Yes 670 45.0 55.0 <0.0001
No, but spoken in home 350 51.5 48.5
Does not speak 5590 62.3 37.7
Aboriginal language and
one is never spoken at home
Member of a Metis cultural
or political organization
No 5350 59.2 40.8 0.0010
Yes 1260 64.3 35.7
Level of spirituality
Very 1500 64.0 36.0 <0.0001
Moderately 3030 60.7 39.3
Not very 1230 58.5 41.5
Not at all 850 54.1 45.9
Gender and age
Women 3440 57.8 42.2 0.0001
Men 3170 62.5 37.5
Age group 18-19 370 64.9 35.1 <0.0001
Age group 20-34 1990 55.8 44.2
Age group 35-49 2340 56.4 43.6
Age group 50-64 1470 65.3 34.7
Age group 65 and older 440 77.3 22.7
Urban/rural geography
CMA 3200 60.9 39.1 0.0251
CA 1300 57.7 42.3
Rural with moderate 950 63.2 36.8
to strong MIZ
Rural with no to weak MIZ 1200 58.3 41.7
Regional geography
Ontario 1420 64.1 35.9 <0.0001
Atlantic 360 69.4 30.6
Quebec 640 59.4 40.6
Prairies 3120 55.4 44.6
British Columbia 1020 66.7 33.3
Territories 70 57.1 42.9
Highest level of education
Less than high school 1460 49.3 50.7 <0.0001
High school or equivalent 1170 56.4 43.6
Some post-secondary or more 3980 65.3 34.7
Household income
Less than $35,000 1630 48.5 51.5 <0.0001
$35,000-$60,899 1670 55.7 44.3
$60,900-$95,899 1650 64.8 35.2
Greater than $95,899 1650 71.5 28.5
Self-perceived health
Excellent 1500 71.3 28.7 <0.0001
Very good 2390 60.7 39.3
Good 1690 53.8 46.2
Fair or poor 1040 52.9 47.1
Binge drinking
No 5060 63.4 36.6 <0.0001
Yes 1560 49.4 50.6
Body mass index
Underweight or normal weight 2450 51.8 48.2 <0.0001
Overweight 2410 63.5 36.5
Obese 1750 73.5 26.5
Leisure-time physical activity
Less than 3 hours/week 3380 56.8 43.2 <0.0001
3 or more hours/week 3230 63.8 36.2
Smoking (current smoker: n = 2640; non-smoker: n = 3970).
Frequency counts for all variables were rounded to the nearest ten,
other than the urban/rural geographic variable, which was rounded
to the nearest fifty.
CMA = Census Metropolitan Area; CA = Census Agglomeration;
MIZ = Metropolitan Influence Zone.
Table 2. Odds ratios relating current smoking to culturally-specific
variables based on the inclusion of various control variables,
Metis 18 years of age and older, Canada (2006 Aboriginal Peoples
Survey and Metis Supplement)
I II
OR (95% CI) OR (95% CI)
Level of spirituality
Very 0.62 * (0.50-0.77) 0.66 * (0.53-0.83)
Moderately 0.73 * (0.60-0.88) 0.75 * (0.61-0.91)
Not very 0.80 * (0.65-0.99) 0.80 * (0.64-0.99)
Not at all ([dagger]) 1.00 (-) 1.00 (-)
Speaks an Aboriginal
language
Yes 2.11 * (1.75-2.56) 2.12 * (1.74-2.58)
No, but spoken in home 1.57 * (1.19-2.05) 1.44 * (1.09-1.90)
Does not speak Aboriginal
language and one is
never spoken at home
([dagger]) 1.00 (-) 1.00 (-)
Last time attending a
Metis cultural event
Less than 1 year ago 1.19 (0.99-1.42) 1.11 (0.92-1.33)
From 1 to 5 years ago 1.17 (0.97-1.40) 1.12 (0.93-1.35)
5 or more years ago 1.12 (0.92-1.33) 1.07 (0.88-1.29)
Never ([dagger]) 1.00 (-) 1.00 (-)
Member of a Metis
organization
Yes 0.76 * (0.65-0.88) 0.76 * (0.65-0.89)
No ([dagger]) 1.00 (-) 1.00 (-)
C-statistic 0.58 0.61
III IV
OR (95% CI) OR (95% CI)
Level of spirituality
Very 0.66 * (0.52-0.83) 0.74 * (0.58-0.93)
Moderately 0.76 * (0.62-0.93) 0.78 * (0.62-0.96)
Not very 0.83 (0.66-1.04) 0.87 (0.69-1.09)
Not at all ([dagger]) 1.00 (-) 1.00 (-)
Speaks an Aboriginal
language
Yes 1.82 * (1.48-2.23) 1.80 * (1.45-2.24)
No, but spoken in home 1.43 * (1.09-1.89) 1.46 * (1.11-1.95)
Does not speak Aboriginal
language and one is
never spoken at home
([dagger]) 1.00 (-) 1.00 (-)
Last time attending a
Metis cultural event
Less than 1 year ago 1.13 (0.93-1.36) 1.17 (0.97-1.42)
From 1 to 5 years ago 1.14 (0.94-1.37) 1.15 (0.95-1.40)
5 or more years ago 1.07 (0.88-1.30) 1.03 (0.85-1.25)
Never ([dagger]) 1.00 (-) 1.00 (-)
Member of a Metis
organization
Yes 0.85 (0.72-1.01) 0.89 (0.75-1.05)
No ([dagger]) 1.00 (-) 1.00 (-)
C-statistic 0.69 0.71
Smoking (current smoker: n = 2640; non-smoker: n = 3970).
Bootstrapped estimates are shown.
I: Culturally-specific variables only; II: controlling for
demographic and geographic variables; III: controlling for
demographic, geographic and socio-economic variables;
IV: controlling for demographic, geographic, socio-economic and
health-related variables.
* Significantly different from reference category (p < 0.05).
([dagger]) Reference category.
