Socio-economic status and smoking in Canada, 1999-2006: has there been any progress on disparities in tobacco use?
Reid, Jessica L. ; Hammond, David ; Driezen, Pete 等
Tobacco use remains the leading cause of preventable death in
Canada, with over 37,000 Canadians dying of tobacco-related illness each
year. (1) This risk behaviour is not randomly distributed within the
population; smoking prevalence exhibits a clear gradient, with lower
socio-economic status groups showing increased rates of tobacco use.
(2-5) Smoking cessation tends to follow an inverse pattern, with higher
socio-economic status (SES) groups more likely to quit smoking. (6-8)
Thus, the burden of tobacco use and related illness is
disproportionately borne by lower SES groups, making tobacco use a
substantial contributor to overall health disparities. (2,9-11) Reducing
smoking, particularly among lower SES populations, is therefore a key
strategy for reducing overall socio-economic inequalities in health and
improving population health. (12)
Population-level policy interventions to reduce tobacco use are
currently being implemented in many countries, and have been credited
with reducing overall smoking prevalence. (13) Canada is a world leader
in tobacco control policies, such as advertising and sale restrictions,
taxation, public smoking bans, and strong antismoking messaging.
However, such policies may have differential effects on subgroups of
smokers, and it is unclear whether the impact of these policies has been
equally distributed. In particular, there is concern that reductions in
smoking prevalence have mainly been achieved among smokers with higher
SES. To date, little information is available on recent Canadian trends
in smoking and quitting by socio-economic status which could provide
insight into tobacco-related disparities and the potential impact of
recent tobacco control efforts on such disparities.
This study examined current smoking trends in Canada, identifying
existing socio-economic disparities and documenting their progress over
time. Specifically, this study sought to identify: 1) any differences by
socio-economic status in smoking prevalence, quit ratios, and rates of
quitting intentions and attempts among smokers, and 2) any changes in
socio-economic differences over time, from 1999 to 2006.
METHODS
This study analyzed data collected through the Canadian Tobacco Use
Monitoring Survey (CTUMS) from 1999 to 2006. CTUMS has been conducted by
Statistics Canada on an ongoing basis since 1999, and was designed to
provide continuous estimates of smoking prevalence in order to monitor
changes over time. The overall design is repeated cross-sectional
surveys of nationally representative samples of Canadians. Data were
collected using computer-assisted telephone interviewing, conducted by
trained interview staff at two Statistics Canada offices. Interviews
were conducted every month, from February to December, in all survey
years. A full description of the CTUMS methodology is available from
Statistics Canada. (14-21)
Sample
The CTUMS target population is all persons 15 years of age and
older, living in Canada, with the exception of residents of the
territories and institutions. The sample design is a two-phase
stratified random sample of telephone numbers, where households are
first selected through random digit dialling, and then individual
respondents are selected based on household composition. An equal number
of respondents in each province are surveyed, and youth (15-24
year-olds) are over-sampled to comprise half of the respondents.
This analysis was limited to respondents 25 years and older
(n=86,971), since education was used as the measure of SES, and is a
less reliable indicator of SES at younger ages when education may still
be in progress.
Measures
Demographic variables included: sex, age (continuous), and region
of residence (Atlantic, Quebec, Ontario, Western, BC). Socioeconomic
status was measured by highest level of education completed, with 4
levels: less than secondary, completed secondary, completed community
college, and completed university. Income information was not available
for all survey years, and thus was not included as a measure of SES.
Smoking status was determined by whether one had smoked 100
cigarettes in their lifetime, current use, and frequency of use
(daily/not). A 3-category smoking status variable classified respondents
into "current smokers" (smoked 100+ cigarettes, currently
smokes daily or occasionally), "former smokers" (smoked 100+
cigarettes, does not currently smoke), and "never-smokers"
(smoked <100 cigarettes, does not currently smoke). From this, a
dichotomous variable for "current smoker" vs.
"non-smoker" (including "former" and
"never") was created. In addition, a smoking frequency
variable was created for all current smokers, distinguishing
"daily" vs. "occasional" smoking. For all smokers,
consumption was measured on a continuous scale of average cigarettes per
day. Respondents were also asked at what age they began smoking, and
total years smoked was calculated for each smoker.
[FIGURE 1 OMITTED]
Quitting-related variables were added to the questionnaire in 2000.
Six-month quit intentions were measured by yes/no responses to the item,
"Are you seriously considering quitting within the next 6
months?" Also, both current smokers and former smokers who had quit
in the past year were asked, "In the past year, how many times did
you stop smoking for at least 24 hours because you were trying to
quit?"
Quit ratios were also calculated for each educational group in each
survey year. The quit ratio was calculated as the number of former
smokers divided by the number of ever-smokers (current and former
smokers) at a given point in time, (22) and provides an indication of
cessation relative to the proportion of smokers within each
socio-economic group.
Analysis
Separate regression models were conducted for each of the dependent
variables to test their associations with education and time. Logistic
regression analysis was conducted using the sample of all respondents
for smoking prevalence (current smoker vs. not), with ever-smokers only
for quit ratios, and with smokers only for smoking frequency (daily
smoker vs. occasional smoker), quit intentions, and quit attempts in the
past year. Multiple regression analysis was conducted using the sample
of smokers for cigarettes per day (CPD) and years smoked. All regression
analyses included age, sex, education level, and region of residence,
and those for quit intentions and quit attempts also controlled for CPD.
Data were analyzed using the survey procedures available in SAS version
9.1 (SAS Institute Inc., Cary, NC) to account for CTUMS' stratified
sampling design using province as the strata variable. Analyses were
also weighted to ensure that estimates are representative of the
Canadian population.
First, the data sets for each year were analyzed separately to
examine any differences between educational groups within the
cross-sectional samples. Second, the full data set was analyzed over
time to test for any time trend in the outcomes. As a final step, the
interaction of time by education was added to these models to examine
whether any associations of education with the outcomes varied over
time.
This research was reviewed by and received ethics clearance from
the Office of Research Ethics at the University of Waterloo, Waterloo,
ON.
RESULTS
Sample
Table 1 shows the demographic characteristics of the full sample
for analysis, including both smokers and non-smokers.
Smoking prevalence
Overall smoking prevalence (including both daily and occasional
smoking) decreased over the study period, from 24% in 1999 to 18% in
2006. This decline was due mainly to decreases in daily smoking, as
occasional smoking prevalence remained stable throughout this time
period. Significant within-year variation was observed by education
level (see Figure 1). Compared to university graduates, respondents with
less than secondary school education had more than 3 times the odds of
being a current smoker, and those who completed secondary school or
college had more than double the odds of being a current smoker (see
Table 2). In the models including all survey years (not shown) there was
a main effect of time ([chi square]=30.3, p<0.0001), but no
interaction between time and education ([chi square]=4.3, p=0.23),
indicating that while prevalence decreased during this time, educational
differences in smoking prevalence were consistent between 1999 and 2006.
Smoking and quitting outcomes for smokers
Table 3 displays the demographic characteristics and outcome
measures for current smokers within each survey year, both overall and
by education level. Each educational group was compared to university
graduates, and significant within-year differences are noted.
Smoking Frequency
The proportion of current smokers reporting daily smoking decreased
from 85% in 1999 to 79% in 2006. Smokers with less education were more
likely to smoke daily (versus occasionally) than university graduates in
almost all survey years, although differences between college and
university groups were not significant in 2000, 2002, and 2006 (see
Table 3). In the models including all years (not shown), there was a
significant effect of time ([chi square]=6.4, p=0.01), but no
significant interaction between time and education ([chi square]=0.28,
p=0.96).
Cigarette Consumption
Average daily cigarette consumption decreased during the study
period, from 16.4 in 1999 to 13.6 in 2006. Cigarette consumption also
differed by education level: university graduates smoked significantly
fewer cigarettes per day than smokers with secondary school education or
less in all survey years except 2000, and significantly less than
college graduates in 2003, 2004, and 2005 (see Table 3). In the models
including all years (not shown), there was a significant effect of time
(F=30.8, p<0.0001), but no significant interaction between time and
education level (F=0.96, p=0.41).
Quit Intentions
Over half of smokers were considering quitting smoking in the next
six months (see Table 3). Few within-year educational differences in
intentions to quit were observed (see Table 3). In the models including
all survey years (not shown), the main effect of time was significant
([chi square]=8.3, p=0.004), but time and education did not interact
([chi square]=3.5, p=0.32).
Quit Attempts
Among current smokers and recent (past year) quitters, nearly half
had made a quit attempt lasting at least 24 hours in the past year (see
Table 3). As with quit intentions, few educational differences were
observed within survey years (see Table 3). In the models including all
years (not shown), the main effect of time was not significant ([chi
square]=0.59, p=0.44), nor was there an interaction between time and
education ([chi square]=5.8, p=0.12).
Quit ratio
The majority of ever-smokers in Canada had quit smoking at the time
they were surveyed (see Table 4). Within each survey year, university
graduates had significantly greater quit ratios than all other groups,
with few exceptions (see Table 4). In the model including all years (not
shown), there was also an overall significant effect of time ([chi
square]=28.4, p<0.0001), but no interaction between education and
time ([chi square]=3.7, p=0.30).
DISCUSSION
Between 1999 and 2006, smoking prevalence and average daily
cigarette consumption declined with similar magnitude in all educational
groups. This translates to over a million fewer smokers in Canada over a
7-year period--a major public health achievement.
However, the findings also highlight persistent socio-economic
disparities in smoking. Canadians with lower education levels were
significantly more likely to smoke: smoking prevalence among the least
educated was approximately double that of the most educated in each year
studied. Although smoking rates generally increased with decreasing
education level, the greatest differences observed were between those
with a university education and all other groups. The patterns and
magnitude of the educational differences in smoking prevalence observed
in this study were similar to those observed in the United States, (23)
although Canadian rates were lower across all groups. Among smokers,
those with lower education were more likely to smoke daily, and the
least educated consumed 3 to 8 more cigarettes per day, on average, than
the most educated. The proportion of ever-smokers who had quit (quit
ratio) also varied considerably by educational group:
university-educated Canadians had the highest quit ratios, indicating
that a greater proportion of smokers in this group have quit. The lack
of interaction between education and time for these outcomes indicates
that educational differences in smoking prevalence and frequency,
cigarette consumption, and quit ratio were stable over the time period
studied.
Studies in other Western countries examining various time periods
over the past two decades have also observed persistent socio-economic
disparities, despite declining prevalence among all groups. (11,24-27)
In the United States, relative educational inequalities in smoking
increased significantly in 40 states between 1990 and 2004,24 although
in another study of overall patterns, the educational gap appears fairly
stable during this time. (11) A study of nine European countries also
found that overall, educational inequalities in smoking prevalence
remained stable among men and increased among women between 1985 and
2000; although, in country-specific analyses, some improvements were
seen in the UK and Italy. (25) On the other hand, an Australian study
found that socio-economic inequalities in smoking remained stable among
women and increased among men between 1989/90 and 2001, (26) and a New
Zealand study found increasing socio-economic inequalities among both
men and women between 1985 and 1996. (27) Overall, the evidence suggests
that socio-economic disparities persist in Western countries, with few
exceptions, and may even be worsening.
The findings for quit intentions and cessation attempts were more
encouraging. Over half of all smokers were considering quitting, and
nearly half had made a quit attempt lasting at least 24 hours in the
past year. Also, few significant educational differences were found in
intentions or attempts to quit, indicating greater socio-economic equity
for these measures.
Limitations
This analysis does, however, face some limitations due to the
nature of secondary analysis. The CTUMS sample did not include those who
live in institutions, on First Nations reserves, or in the
territories--all groups with higher smoking rates and generally lower
SES. The sample was also limited to those over 25, and some studies have
found the greatest socio-economic disparities among younger birth
cohorts, (7,28) so estimates for within-year SES differences may be
conservative. Also, some data were not available in all survey years due
to variation in questionnaire content (e.g., quit attempts). In
addition, education was the sole measure of SES; however, previous
research has identified education as the SES variable with the greatest
correlation with smoking. (7,22) Finally, the cross-sectional nature of
the survey meant that individual smoking trajectories and cessation
outcomes could not be followed and assessed, which would have
strengthened the conclusions made and allowed richer analysis.
CONCLUSIONS AND IMPLICATIONS
The decline in smoking prevalence among Canadians in all education
groups between 1999 and 2006 represents a major public health
achievement. However, considerable smoking-related disparities persist
among socio-economic groups, and have changed very little in the last
decade as the tobacco control landscape in Canada has evolved.
Therefore, while recent programs and policies have succeeded in reducing
tobacco use in all groups, they have not impacted the disparities
between socio-economic groups. Increasing access to existing tobacco
interventions and/or designing specific, targeted interventions may
benefit socio-economically disadvantaged smokers, who are
disproportionately affected by the harms of tobacco use.
Acknowledgements: This study analyzed public-use data collected by
Statistics Canada. The results and views expressed in this paper are
those of the authors and not Statistics Canada. The first author was
supported by a Canadian Institutes of Health Research (CIHR) Strategic
Training Program in Tobacco Research (STPTR) Fellowship and an Ontario
Graduate Scholarship. Additional support was provided by the Centre for
Behavioural Research and Program Evaluation, National Cancer Institute
of Canada/Canadian Cancer Society. The authors gratefully acknowledge
the statistical advice of Christian Boudreau and Mary Thompson.
Received: January 7, 2009 Accepted: May 5, 2009
REFERENCES
(1.) Baliunas D, Patra J, Rehm J, Popova S, Kaiserman M, Taylor B.
Smoking-attributable mortality and expected years of life lost in Canada
2002: Conclusions for prevention and policy. Chron Dis Can
2007;27(4):154-62.
(2.) Jarvis MJ, Wardle J. Social patterning of individual health
behaviours: The case of cigarette smoking. In: Marmot M, Wilkinson RG
(Eds.), Social Determinants of Health, Second Ed. Oxford, UK: Oxford
University Press, 2006.
(3.) Bobak M, Jha P, Nguyen S, Jarvis M. Poverty and smoking. In:
Jha P, Chaloupka FJ (Eds.), Tobacco Control in Developing Countries. New
York, NY: Oxford University Press, 2000.
(4.) US Department of Health and Human Services. Women and Smoking:
A Report of the Surgeon General. Atlanta, GA: US Department of Health
and Human Services, Public Health Service, Centers for Disease Control,
National Center for Chronic Disease Prevention and Health Promotion,
Office on Smoking and Health, 2001.
(5.) US Department of Health and Human Services. Tobacco Use Among
U.S. Racial/Ethnic Minority Groups--African Americans, American Indians
and Alaska Natives, Asian Americans and Pacific Islanders, and
Hispanics: A Report of the Surgeon General. Atlanta: U.S. Department of
Health and Human Services, Centers for Disease Control and Prevention,
National Center for Chronic Disease Prevention and Health Promotion,
Office on Smoking and Health, 1998.
(6.) Kunst A, Giskes K, Mackenbach J. Socioeconomic Inequalities in
Smoking in the European Union: Applying an Equity Lens to
Tobacco-control Policies. Brussels: European Network of Smoking
Prevention, 2006.
(7.) Schaap MM, Kunst AE, Leinsalu M, Regidor E, Ekholm O, Dzurova
D, et al. Effect of nationwide tobacco control policies on smoking
cessation in high and low educated groups in 18 European countries. Tob
Control 2008;17:248-55.
(8.) Gilpin EA, Pierce JP. Demographic differences in patterns of
smoking cessation: United States, 1950-1990. Ann Epidemiol
2002;12(3):141-50.
(9.) Jha P, Peto R, Zatonski W, Boreham J, Jarvis MJ, Lopez AD.
Social inequalities in male mortality, and in male mortality from
smoking: Indirect estimation from national death rates in England and
Wales, Poland, and North America. Lancet 2006;368(9533):367-70.
(10.) Siahpush M, English D, Powles J. The contribution of smoking
to socioeconomic differentials in mortality: Results from the Melbourne
Collaborative Cohort Study, Australia. J Epidemiol Community Health
2006;60:1077-79.
(11.) Meara ER, Richards S, Cutler DM. The gap gets bigger: Changes
in mortality and life expectancy, by education, 1981-2000. Health Aff
2008;27(2):350-60.
(12.) Ogilvie D, Petticrew M. Reducing social inequalities in
smoking: Can evidence inform policy? A pilot study. Tob Control
2004;13:129-31.
(13.) World Health Organization. WHO Report on the Global Tobacco
Epidemic, 2008: The MPOWER Package. Geneva, Switzerland: WHO, 2008.
(14.) Statistics Canada. Microdata User Guide: Canadian Tobacco Use
Monitoring Survey, Annual, February - December 2006. Available at:
http://www.statcan.gc.ca/dli-ild/meta/ctums-esutc/2006/
annual-annuel/2006-ctums-estuc-anng-eng.pdf (Accessed January 9, 2008).
(15.) Statistics Canada. Microdata User Guide: Canadian Tobacco Use
Monitoring Survey, Annual, February - December 2005. Available at:
http://www.statcan.gc.ca/dli-ild/meta/ctums-esutc/2005/
annual-annuel/5902728-eng.pdf (Accessed January 9, 2008).
(16.) Statistics Canada. Microdata User Guide: Canadian Tobacco Use
Monitoring Survey, Annual, February-December 2004. Available at:
http://www.statcan.gc.ca/dli-ild/meta/ctums-esutc/2004/
annual-annuel/2004-ctums-estuc-anng-eng.pdf (Accessed January 9, 2008).
(17.) Statistics Canada. Microdata User Guide: Canadian Tobacco Use
Monitoring Survey, Annual, February-December 2003. Available at:
http://www.statcan.gc.ca/dli-ild/meta/ctums-esutc/2003/
annual-annuel/4187682-eng.pdf (Accessed January 9, 2008).
(18.) Statistics Canada. Microdata User Guide: Canadian Tobacco Use
Monitoring Survey, Annual, February-December 2002. Available at:
http://www.statcan.gc.ca/dli-ild/meta/ctums-esutc/2002/4215775-eng.pdf
(Accessed January 9, 2008).
(19.) Statistics Canada. Microdata User Guide: Canadian Tobacco Use
Monitoring Survey, Annual, February-December 2001. Available at:
http://www.statcan.gc.ca/dli-ild/meta/ctums-esutc/2001/
annual-annuel/4187655-eng.pdf (Accessed January 9, 2008).
(20.) Statistics Canada. Microdata User Guide: Canadian Tobacco Use
Monitoring Survey, Annual, February-December 2000. Available at:
http://www.statcan.gc.ca/dli-ild/meta/ctums-esutc/2000/
annual-annuel/4187646-eng.pdf (Accessed January 9, 2008).
(21.) Statistics Canada. Microdata User Guide: Canadian Tobacco Use
Monitoring Survey, Annual, February-December 1999. Available at:
http://www.statcan.gc.ca/dli-ild/meta/ctums-esutc/1999/ annual-annuel/
4216924-eng.pdf (Accessed January 9, 2008).
(22.) US Department of Health and Human Services. Reducing the
Health Consequences of Smoking--25 Years of Progress: A Report of the
Surgeon General. Atlanta: US Department of Health and Human Services,
Public Health Service, Centers for Disease Control, National Center for
Chronic Disease Prevention and Health Promotion, Office on Smoking and
Health, 1989.
(23.) Fagan P, Moolchan E, Lawrence D, Fernander A, Ponder PK.
Identifying health disparities across the tobacco continuum. Addiction
2007;102(Suppl 2):5-29.
(24.) Harper S, Lynch J. Trends in socioeconomic inequalities in
adult health behaviors among US states, 1990-2004. Public Health Rep
2007;122:177-89.
(25.) Giskes K, Kunst AE, Benach J, Borrell C, Costa G, Dahl E, et
al. Trends in smoking behaviour between 1985 and 2000 in nine European
countries by education. J Epidemiol Community Health 2005;9:395-401.
(26.) Najman JM, Toloo G, Siskind V. Socioeconomic disadvantage and
changes in health risk behaviours in Australia: 1989-1990 to 2001. Bull
WHO 2006;84:12.
(27.) Hill SE, Blakely TA, Fawcett JM, Howden-Chapman P. Could
mainstream antismoking programs increase inequalities in tobacco use?
New Zealand data from 1981-96. Aust N Z J Public Health 2005;29:279-84.
(28.) Cavelaars AEJM, Kunst AE, Geurts JJM, Crialesi R, Grotvedt L,
Helmert U, et al. Educational differences in smoking: International
comparison. BMJ 2000;320:1102-7.
Jessica L. Reid, MSc, David Hammond, PhD, Pete Driezen, MSc
Authors' Affiliation
Department of Health Studies & Gerontology, University of
Waterloo, Waterloo, ON
Correspondence and reprint requests: Jessica L. Reid, Department of
Health Studies & Gerontology, University of Waterloo, 200 University
Ave. W., Waterloo, ON N2L 3G1, Tel: 519-888-4567, ext. 38549, Fax:
519-746-2510, E-mail: jl3reid@uwaterloo.ca
Table 1. Sample Characteristics, by Survey Year, 1999-2006 (n=86,971)
Survey Year
1999 2000 2001
Variable (n=10,634) (n=10,154) (n=10,958)
Sex
Male 44.2 44.2 44.2
Female 55.8 55.8 55.8
Age (range: 25-85)
Mean 48.3 48.6 48.9
SD 15.6 15.5 15.7
Education level
Less than secondary 26.7 26.4 24.9
Completed secondary 39.4 40.0 40.8
Completed college 14.4 14.6 15.1
Completed university 19.5 19.1 19.2
Region
Atlantic 40.6 41.3 39.6
Quebec 9.4 10.0 10.1
Ontario 8.8 8.0 10.3
Western 31.7 30.8 29.7
British Columbia 9.4 10.0 10.4
Survey Year
2002 2003 2004
Variable (n=11,855) (n=10,849) (n=10,404)
Sex
Male 43.5 44.0 44.2
Female 56.5 56.0 55.8
Age (range: 25-85)
Mean 49.6 49.9 49.7
SD 15.4 15.6 15.5
Education level
Less than secondary 25.6 24.0 21.0
Completed secondary 37.2 37.8 38.4
Completed college 17.0 17.9 17.2
Completed university 20.2 20.3 23.5
Region
Atlantic 37.2 37.8 39.9
Quebec 9.2 9.7 10.1
Ontario 18.2 9.7 9.5
Western 26.2 27.5 31.0
British Columbia 9.2 15.3 9.5
Survey Year
2005 2006
Variable (n=10,797) (n=11,320)
Sex
Male 43.3 44.3
Female 56.7 55.7
Age (range: 25-85)
Mean 50.6 50.8
SD 15.6 15.6
Education level
Less than secondary 21.2 20.6
Completed secondary 36.7 36.2
Completed college 18.8 17.7
Completed university 23.4 25.5
Region
Atlantic 36.2 40.9
Quebec 8.6 9.7
Ontario 8.4 9.2
Western 38.4 30.6
British Columbia 8.5 9.6
Data presented as unweighted percentages unless otherwise noted
Table 2. Odds Ratios for Current Smoking (Compared to University
Graduates), 1999-2006
Survey Year
1999 2000 2001
Education Level (n=10,634) (n=10,154) (n=10,958)
Less than secondary 2.95 * 4.05 * 3.46 *
(2.24-3.88) (3.05-5.38) (2.64-4.53)
Completed secondary 1.92 * 2.50 * 2.37 *
(1.49-2.46) (1.96-3.20) (1.86-3.02)
Completed college 1.26 2.01 * 1.81 *
(0.94-1.71) (1.48-2.73) (1.35-2.42)
Completed university ref. ref. ref.
Survey Year
2002 2003 2004
Education Level (n=11,855) (n=10,849) (n=10,404)
Less than secondary 3.25 * 3.20 * 3.18 *
(2.54-4.14) (2.38-4.30) (2.36-4.30)
Completed secondary 2.26 * 2.36 * 2.09 *
(1.83-2.79) (1.81-3.07) (1.63-2.67)
Completed college 1.69 * 1.86 * 1.59 *
(1.33-2.15) (1.38-2.50) (1.18-2.14)
Completed university ref. ref. ref.
Survey Year
2005 2006
Education Level (n=10,797) (n=11,320)
Less than secondary 3.77 * 3.79 *
(2.73-5.21) (2.81-5.11)
Completed secondary 2.85 * 2.62 *
(2.20-3.68) (2.07-3.31)
Completed college 2.00 * 1.96 *
(1.49-2.68) (1.48-2.59)
Completed university ref. ref.
Data presented as ORs (95% confidence intervals), from weighted
regression models conducted separately for each year, controlling
for age, sex, and region
* Significantly different (at p<0.01) from university graduate
education level, in weighted regression models conducted
separately for each year that controlled for age, sex, and region
Table 3. Characteristics of Smokers, by Education Level and Survey
Year, 1999-2006 (n=19,033)
Survey Year
1999 2000
Variable, Education Level (n=2705) (n=2583)
Sex (% male) 52.9% 52.1%
Age (years) 43.5 (13.1) 43.9 (13.1)
Daily smoking (%)
All groups 85.0% 82.8%
Less than secondary 90.4 ** 87.1 **
Completed secondary 86.5 ** 86.5 **
Completed college 85.5 * 75.4
Completed university (ref) 71.9 70.6
Cigarettes per day
All groups 16.4 (10.2) 15.6 (10.4)
Less than secondary 19.2 ** (10.6) 18.2 (11.1)
Completed secondary 16.0 * (9.2) 15.5 (8.8)
Completed college 15.2 (9.4) 12.9 (9.1)
Completed university (ref) 13.6 (11.8) 14.6 (14.2)
Years smoked
All groups 27.3 (12.8) 28.2 (13.3)
Less than secondary 32.2 (13.7) 34.6** (15.0)
Completed secondary 26.5 * (11.7) 26.4 (11.7)
Completed college 23.8 (11.4) 24.5 (11.3)
Completed university (ref) 25.0 (13.4) 25.8 (12.6)
Intending to quitt (%)
All groups -- 53.6%
Less than secondary -- 44.4
Completed secondary -- 57.3
Completed college -- 53.7
Completed university (ref) -- 59.3
Made a quit attemptt (%)
All groups -- --
Less than secondary -- --
Completed secondary -- --
Completed college -- --
Completed university (ref) -- --
Survey Year
2001 2002
Variable, Education Level (n=2542) (n=2471)
Sex (% male) 54.4% 53.5%
Age (years) 44.1 (13.2) 43.7 (13.1)
Daily smoking (%)
All groups 85.5% 84.0%
Less than secondary 91.4 ** 91.5 **
Completed secondary 83.7 * 85.8 **
Completed college 90.2 ** 76.5
Completed university (ref) 75.6 74.6
Cigarettes per day
All groups 15.1 (9.5) 15.3 (9.6)
Less than secondary 18.2 ** (11.0) 18.2 ** (10.6)
Completed secondary 15.0 ** (9.0) 15.6 ** (8.7)
Completed college 13.5 (8.0) 12.7 (9.3)
Completed university (ref) 11.8 (11.8) 12.4 (9.0)
Years smoked
All groups 28.0 (13.0) 28.0 (12.9)
Less than secondary 33.9 ** (14.9) 32.7 ** (13.3)
Completed secondary 27.4 ** (12.4) 27.8 ** (12.1)
Completed college 23.9 (10.2) 23.1 (11.9)
Completed university (ref) 25.0 (10.9) 26.8 (12.9)
Intending to quitt (%)
All groups 56.7% 59.8%
Less than secondary 50.6 ** 53.9
Completed secondary 53.6 ** 59.7
Completed college 62.7 66.4
Completed university (ref) 70.2 61.9
Made a quit attemptt (%)
All groups -- --
Less than secondary -- --
Completed secondary -- --
Completed college -- --
Completed university (ref) -- --
Survey Year
2003 2004
Variable, Education Level (n=2276) (n=2095)
Sex (% male) 55.9% 55.4%
Age (years) 43.7 (12.9) 42.9 (12.7)
Daily smoking (%)
All groups 82.6% 78.8%
Less than secondary 86.2 ** 88.1 **
Completed secondary 86.2 ** 81.6 **
Completed college 81.5 * 79.5 **
Completed university (ref) 68.5 62.7
Cigarettes per day
All groups 14.6 (9.3) 13.6 (9.4)
Less than secondary 17.1 ** (10.2) 17.3 ** (9.4)
Completed secondary 15.0 ** (8.6) 13.9 ** (9.7)
Completed college 13.5 * (8.6) 13.1 ** (8.6)
Completed university (ref) 11.2 (9.6) 9.5 (7.9)
Years smoked
All groups 28.4 (13.0) 27.2 (12.9)
Less than secondary 34.1 (13.8) 33.2 ** (13.7)
Completed secondary 27.2 (12.6) 27.1 ** (12.7)
Completed college 25.6 (10.5) 25.8 * (10.7)
Completed university (ref) 26.5 (13.3) 22.9 (12.4)
Intending to quitt (%)
All groups 56.7% 58.0%
Less than secondary 52.3 51.9
Completed secondary 56.4 61.1 *
Completed college 60.8 65.7 *
Completed university (ref) 58.6 49.6
Made a quit attemptt (%)
All groups -- 44.5%
Less than secondary -- 39.9
Completed secondary -- 47.7
Completed college -- 46.0
Completed university (ref) -- 40.5
Survey Year
2005 2006
Variable, Education Level (n=2109) (n=2252)
Sex (% male) 57.2% 52.9%
Age (years) 43.5 (12.4) 44.2 (12.9)
Daily smoking (%)
All groups 83.4% 79.1%
Less than secondary 90.4 ** 86.1 **
Completed secondary 84.4 ** 83.4 **
Completed college 88.2 ** 74.1
Completed university (ref) 67.1 67.3
Cigarettes per day
All groups 14.6 (9.6) 13.6 (9.4)
Less than secondary 17.3 ** (9.7) 14.7 (10.7)
Completed secondary 15.6 ** (9.7) 14.7 * (9.1)
Completed college 13.2 * (7.7) 11.6 (7.7)
Completed university (ref) 10.6 (9.8) 12.0 (9.8)
Years smoked
All groups 27.6 (12.7) 28.5 (12.9)
Less than secondary 32.7 ** (14.8) 33.5 ** (13.9)
Completed secondary 27.5 * (11.8) 28.8 ** (12.8)
Completed college 24.9 (11.7) 24.8 * (9.9)
Completed university (ref) 25.3 (11.7) 26.4 (13.1)
Intending to quitt (%)
All groups 57.9% 65.7%
Less than secondary 50.7 60.4
Completed secondary 55.6 65.1
Completed college 65.4 69.4
Completed university (ref) 63.1 68.5
Made a quit attemptt (%)
All groups 47.2% 46.2%
Less than secondary 52.4 * 43.4
Completed secondary 43.8 45.4
Completed college 55.8 * 52.1
Completed university (ref) 40.2 45.2
Data presented as weighted means (standard deviation) unless
otherwise noted
* Significantly different (at p<0.05) from university graduate
education level, in weighted regression models (logistic
regression for daily smoking and quit intentions and attempts;
multiple regression for CPD and years smoked) conducted
separately for each year that controlled for age, sex, and region
(and CPD for quit intentions and attempts analyses)
** Significantly different (at p<0.01) from university graduate
education level, in weighted regression models (logistic
regression for daily smoking and quit intentions and attempts;
multiple regression for CPD and years smoked) conducted
separately for each year that controlled for age, sex, and
region (and CPD for quit intentions and attempts analyses)
([dagger]) "Seriously considering quitting smoking within the
next 6 months"; added to the survey in 2000
([double dagger]) Made a quit attempt lasting at least 24 hours
within the past year; added to the survey in 2000, but analyzed only
for 2004 through 2006, due to inconsistencies in question
coverage between this period and earlier survey waves (e.g.,
2000-2002 asked current smokers; 2003 included only smokers who
had tried to quit in the past 2 years; 2004-2006 asked current
smokers and former smokers who had quit in past 12 months)
Table 4. Percentage of Ever-Smokers Who Have Quit (Quit Ratio), by
Education Level and Year, 1999-2006
Survey Year
1999 2000 2001
Education Level (n=5859) (n=5548) (n=5799)
Less than secondary 52.6 % ** 50.4 % ** 55.8 % **
(48.3-56.9) (45.8-55.0) (51.6-60.0)
Completed secondary 52.6 % ** 53.3 % ** 53.5 % **
(49.1-56.2) (49.7-57.0) (50.0-56.9)
Completed college 55.60% 52.8 % ** 53.3 % **
(49.7-61.6) (46.5-59.1) (47.6-59.0)
Completed university (ref) 61.7 % 69.3 % 68.5 %
(56.0-67.5) (64.1-74.5) (63.5-73.6)
Survey Year
2002 2003 2004
Education Level (n=6224) (n=5866) (n=5427)
Less than secondary 57.6 % ** 60.8 % ** 60.1 % **
(53.8-61.3) (56.4-65.1) (55.2-65.1)
Completed secondary 55.7 % ** 57.9 % * 58.9 % **
(52.8-58.7) (54.5-61.3) (55.3-62.5)
Completed college 58.8 % 57.3 % 59.0 %
(54.3-63.2) (52.1-62.6) (53.4-64.5)
Completed university (ref) 65.3 % 65.7 % 66.1 %
(60.6-70.1) (60.0-71.5) (60.8-71.5)
Survey Year
2005 2006
Education Level (n=5654) (n=6048)
Less than secondary 64.5 % ** 63.7 % **
(59.2-69.9) (59.0-68.5)
Completed secondary 59.4 % ** 60.6 % **
(55.5-63.3) (57.2-64.0)
Completed college 60.1 % ** 62.6 %
(54.7-65.5) (57.5-67.7)
Completed university (ref) 74.6 % 69.4 %
(70.3-78.9) (64.9-73.9)
Data presented as weighted percentages (95% confidence limits)
* Significantly different (at p<0.05) from university graduate
education level, in weighted logistic regression models
conducted separately for each year that controlled for age,
sex, and region
** Significantly different (at p<0.01) from university
graduate education level, in weighted logistic regression
models conducted separately for each year that controlled for age,
sex, and region
