Article Text
PDF
Abstract
Objective To explore between-country differences and within-country trends over time in smokers' reasons for thinking about quitting and the relationship between reasons and making a quit attempt.
Methods Participants were nationally representative samples of adult smokers from the UK (N=4717), Canada (N=4884), the USA (N=6703) and Australia (N=4482), surveyed as part of the International Tobacco Control Four Country Survey between 2002 and 2015. Generalised estimating equations were used to evaluate differences among countries in smokers' reasons for thinking about quitting and their association with making a quit attempt at follow-up wave.
Results Smokers' concern for personal health was consistently the most frequently endorsed reason for thinking about quitting in each country and across waves, and was most strongly associated with making a quit attempt. UK smokers were less likely than their counterparts to endorse health concerns, but were more likely to endorse medication and quitline availability reasons. Canadian smokers endorsed the most reasons, and smokers in the USA and Australia increased in number of reasons endorsed over the course of the study period. Endorsement of health warnings, and perhaps price, appears to peak in the year or so after the change is introduced, whereas other responses were not immediately linked to policy changes.
Conclusions Differences in reasons for thinking about quitting exist among smokers in countries with different histories of tobacco control policies. Health concern is consistently the most common reason for quitting and the strongest predictor of future attempts.
- Public policy
- Packaging and Labelling
- Price
- Media
Statistics from Altmetric.com
Introduction
Numerous studies using various samples of current and former cigarette smokers identify health concerns as the most common reason for wanting to quit smoking.1 Other commonly cited reasons for quitting include social concerns, such as modelling behaviour for children or pressure from others to quit, and cost, but prevalence of endorsement of these reasons is typically modest in comparison to prevalence of health concerns. Hummel et al 2 evaluated trends in thinking about quitting in six European countries and found that cigarette price increases and warning labels trigger thoughts about quitting. Absent from the literature, however, is a comprehensive evaluation among the UK, Canada, the USA and Australia, in smokers' reasons for thinking about quitting situated within the context of the tobacco control policies that were implemented to directly or indirectly impact quit attempt rates in these countries.
Table 1 presents an overview of tobacco control policies that were implemented in the UK, Canada, the USA and Australia between 2001 and 2015. The UK's national tobacco control strategy, first laid out in 1998 with the publication, Smoking Kills: A White Paper on Tobacco,3 included a commitment for significant increased government investment in media campaigns, increases in tobacco price, reduction in smoking in public places, and increased investment in intensive resources for smoking cessation (including free or subsidised stop-smoking medications and face-to-face behavioural support from trained specialists). Since then, mass media expenditure in the UK rose and peaked in 2004/2005, declined and dropped to £0 in 2010/2011, then rose until 2012/2013 before starting to decline again. Annual tobacco tax increases were at or slightly above inflation rates until 2010, and funding on cessation services in the UK has been declining steadily over recent years with some services closing. During the past 15 years, Canada added pictorial health warnings to cigarette packs and introduced a revised set of larger pictorial warnings covering 75% of the pack in 2012; other policies implemented in Canada included bans on tobacco advertising, bans on smoking in indoor public places and bans on point-of-sale tobacco displays. During the same time period, the USA implemented a national media antismoking campaign direct at youth, created a national network of telephone helplines and raised the federal excise tax on cigarettes; many states in the USA also raised cigarette taxes and adopted clean indoor air laws. Australia has been funding a comprehensive national mass media antismoking campaign aimed at raising public awareness of the health risks of smoking since 1997 (Australian National Tobacco Campaign); other policies implemented in Australia included bans on terms like light and mild, augmenting cigarette pack warnings with graphic images in 2006, banning smoking in licensed venues in 2006 and 2007 (state by state), a 25% increase in tax in 2010, and both increasing warning size and mandating standardised packaging in 2012.
Tobacco control policies in the UK, Canada, the USA and Australia between 2001 and 2015*
The International Tobacco Control Four Country Survey (ITC-4C) asks smokers from the UK, Canada, the USA and Australia, about their reasons for thinking about quitting, over the time period when many of these tobacco control policies were implemented in each country. The longitudinal design of the ITC-4C further allows us to evaluate the association between thinking about quitting for a variety of reasons and making a quit attempt at follow-up survey. Within the context of the tobacco control policies implemented in each country, the purpose of this study was to explore (1) between-country differences in smokers' reasons for thinking about quitting, (2) within-country trends in smokers' reasons for thinking about quitting, and (3) associations between reasons for thinking about quitting and making a quit attempt at follow-up wave.
Methods
Participants
Participants were adults aged 18+ from the UK, Canada, the USA and Australia, who were interviewed as part of the ITC-4C between 2002 and 2015. The ITC-4C is a prospective cohort survey that used random digit dialling to recruit nationally representative samples of ∼2000 smokers from each country in 2002. Response rates ranged from 26% (USA) to 50% (Canada), and prior analyses have demonstrated that the demographic profiles of respondents to this survey were similar to the profiles of those who participated in national benchmark surveys, suggesting that any non-response to this survey is comparable to that of benchmark surveys.4 Participants were recontacted approximately annually to complete follow-up surveys and new smokers were recruited each year to offset those lost to attrition (∼30% on average5). Smokers who subsequently quit smoking were retained in the sample. The study protocol was approved by the Institutional Review Boards/Research Ethics Boards of the University of Waterloo (Canada), Roswell Park Cancer Institute (USA), University of Strathclyde (UK), University of Stirling (UK), The Open University (UK) and The Cancer Council Victoria (Australia). Detailed descriptions of the ITC-4C survey have been published elsewhere.4–7
All participants were current smokers at recruitment into the study (ie, they smoked at least 100 cigarettes in their lifetimes and smoked at least 1 cigarette in the past 30 days). Evaluation of between-country differences and within-country trends in reasons for thinking about quitting included current daily smokers at each wave (N=20 786 individuals and N=47 590 observations), and evaluation of making a quit attempt included current daily smokers at previous wave (N=14 071 individuals and N=18 526 observations).
Measures
Reasons for thinking about quitting
During each survey wave, smokers were asked about various reasons for thinking about quitting: concern for personal health; concern about the effect of cigarette smoke on non-smokers; society disapproves of smoking; price of cigarettes; smoking restrictions at work; smoking restrictions in public places; advice from a doctor, dentist or other health professional to quit; free or lower cost stop-smoking medication (SSM); availability of telephone helpline/quitline/information line; advertisements or information about the health risks of smoking; warning labels on cigarette packages; setting an example for children. Respondents were asked each item separately, and response options were: not at all, somewhat or very much, which were dichotomised to indicate whether a reason was endorsed at all (ie, somewhat or very much) versus not at all.
Owing to the high correlations among reasons, we performed exploratory factor analysis so we could organise our presentation of reasons according to their relatedness. Concern for personal health, setting an example for children and concern about effect on non-smokers were highly related so we grouped them as ‘concern/risk’ reasons; smoking restrictions at work, smoking restrictions in public places and society disapproves of smoking were highly related and were grouped as ‘public/society’ reasons; free/lower cost stop-smoking medication, availability of telephone helpline/quitline/information line, and doctor/dentist/other health professional advice to quit were related and were grouped as ‘assistance’ reasons; advertisements/information about health risks and warning labels on cigarette packs were moderately related and were grouped as ‘information’ reasons, even though they were also somewhat related to the ‘concern/risk’ reasons; price of cigarettes stood out as being least correlated with the other reasons (see online supplementary table S1). We computed the average per cent endorsement for each group of related reasons using predicted values from adjusted regression models as described below in the Analysis section. We also calculated the total number of reasons endorsed in each country.
supplementary table
Factor analysis of reasons to think about quitting
Nicotine dependence
During each survey, smokers were asked how many cigarettes they smoke per day (CPD), categorised into 1–10, 11–20, 21–30 and 31+, and how soon after waking they have the first cigarette (time to first cigarette, TTFC), categorised into >60, 31–60, 6–30 and 0–5 min.
Making a quit attempt
During each follow-up survey, prior year smokers were asked, ‘Have you made any attempts to stop smoking since we last talked with you?’
Demographics and other covariates
The following covariates were included in adjusted analyses: sex, age group (ie, 18–29, 30–54, 55+), majority/minority group (based on the standard way of identifying minorities in each country, ie, racial/ethnic group in the UK, Canada and the USA and English language spoken at home in Australia), level of education (ie, ‘low’ if completed high school or less in Australia, Canada and the USA, or secondary/vocational or less in the UK, ‘moderate’ if completed college/university (no degree) in the UK, technical/trade/some university (no degree) in Australia, or community college/trade/technical school/some university (no degree) in Canada and the USA, or ‘high’ if completed university or postgraduate in all countries), annual household income (defined as ‘low’ if <US$30 000 (Australia, Canada, USA) or <£30 000 (UK), ‘moderate’ if between US$30 000 and US$59 999 (or £30 000 and £44 999 in the UK), or ‘high’ if ≥US$60 000 (or £45 000 in the UK), time in sample (ie, the number of waves a respondent participated in the ITC survey) and difference in time between survey waves (because the gap in time between surveys differed by country at the end of the study period). The exact wording of all items used in the ITC surveys can be found at: http://www.itcproject.org.8
Statistical analyses
Generalised estimating equations (GEEs) were used so that participants from all waves of the study period could be included in analyses at once while statistically controlling for dependence among observations from the same individuals.9 ,10 Regression models specified the unstructured within-person correlation matrix and CIs were calculated using a robust variance estimator. For each analysis, model covariance parameters were set at a maximum of 100 iterations and convergence tolerance for the coefficient vector was set at 1e-6. All analyses were conducted using Stata V.11 (Stata Statistical Software, Version 11. College Station, TX: StataCorp LP, 2009).
Between-country differences in reasons for thinking about quitting
Among current daily smokers, regression analyses were used to model the association between country and endorsement of reasons, aggregated across the study period and adjusted for sex, age group, majority/minority group, income, education, CPD, TTFC, survey wave, time in sample and difference in time between waves, which are hereafter referred to as ‘covariates’. For each country, post hoc estimation was used to generate predicted values of endorsement of each reason and each group of reasons.
Within-country trends in reasons for thinking about quitting
Among current daily smokers, prevalence of endorsement of reasons was plotted by country and by wave, and linear, quadratic and cubic trends in endorsement of reasons over the course of the study period were evaluated using logistic regression analyses (adjusted for covariates). That is, wave of the study period was the independent variable and endorsement of reasons was the dependent variable; ORs for the linear trends indicate the odds of endorsing each reason per 1 unit increase in wave. When significant non-linear trends were found, we reviewed the prevalence estimates plotted over time alongside the policy changes that took place over time and ran time-limited regression analyses to test whether specific policy changes are associated with changes in endorsement of reasons.
Reasons for thinking about quitting and making a quit attempt
Among current daily smokers at previous wave, regression analyses were used to evaluate the association between the reasons and making a quit attempt, aggregated across the study period and across the four countries. One set of models was adjusted for country and the covariates (model 1), and a second set of models was adjusted for country, the covariates and each other reason (model 2). For each set of models, interactions between country and each reason were tested.
Results
Between-country differences in reasons for thinking about quitting
Table 2 shows country differences in reasons for thinking about quitting aggregated across the study period. The most frequently endorsed individual reasons for thinking about quitting were concern for personal health, price of cigarettes and setting an example for children. Smokers in the UK or Canada were more likely to endorse ‘public/society’ reasons (42.5% and 46.4%, respectively) than smokers in the USA (37.0%) or Australia (39.1%). Smokers in the UK or the USA were more likely to endorse ‘assistance’ reasons (39.6% for each country) than smokers in Canada (38.5%) or Australia (37.1%). Canadian smokers were most likely to endorse ‘concern/risk’ reasons (73.8%), ‘public/society’ reasons (46.4%) and ‘information’ reasons (41.3%); Canadian smokers also endorsed the greatest number of reasons (mean=6.3 out of 12). Smokers in the USA were most likely to endorse ‘price’ as a reason for thinking about quitting (74.5%).
Endorsement of reasons that led smokers to think about quitting, by country
Within-country trends in reasons for thinking about quitting
Table 3 shows within-country change in endorsement of reasons for thinking about quitting per wave, and online supplementary figure S1 shows these data plotted at each wave of the study period for each country. Several reasons were increasingly reported over the 13 years, but there were nearly as many non-linear changes.
Linear trends in endorsement of reasons that led smokers to think about quitting, by country
supplementary figure
Trends in endorsement of reasons that led daily smokers to think about quitting, by country. Estimates adjusted for sex, age group, majority/minority group, education, income, CPD, TTFC, survey wave, time in sample, and difference in time between waves.
Upward trends
In the UK, there were upward trends in endorsement of smoking restrictions in public places and at work. In Canada, there was an overall upward trend for the group of ‘assistance’ reasons, with an individual upward trend for availability of free/lower cost SSMs. In the USA, there were overall upward trends for the sum of reasons endorsed, the group of ‘public/society’ reasons, and the group of ‘assistance’ reasons, with individual upward trends for smoking restrictions at work, availability of free/lower cost SSMs, availability of a telephone helpline/quitline and advice from a health professional. In Australia, there were overall upward trends for the sum of reasons endorsed, the group of ‘assistance’ reasons and the group of ‘information’ reasons, with individual upward trends for smoking restrictions at work, availability of free/lower cost SSMs, availability of a telephone helpline/quitline, advice from a health professional and warning labels on cigarette packs. Specific policy-related upward trends tended to occur in countries where relevant policies were implemented during the middle to end of the study period.
Downward trends
In the UK, there was an overall downward trend for the group of ‘concern/risk’ reasons and the group of ‘information’ reasons, with individual downward trends for concern for health of others, advertisements about health risks and warning labels on cigarette packs. In Canada, there was an overall downward trend for the group of ‘information’ reasons with individual downward trends for advertisements about health risks, warning labels on cigarette packs and the price of cigarettes. These policy-related downward trends tended to occur where relevant policies were implemented during the beginning of the study period. There were no downward trends in the USA or Australia.
Non-linear trends
Various non-linear changes in endorsement of reasons were observed in different countries over the course of the study period, as identified in table 3 and depicted in online supplementary figure S1. Some of these shifts in reasons align with related policy changes; in Australia, the rise in endorsing warning labels as a reason between 2005/2006 and 2006/2007, and again between 2010/2011 and 2013/2015, corresponds with when new stronger warnings were added, and in the latter case, standardised packaging (adjusted OR (AOR)=2.20, p<0.001 and AOR=1.56, p<0.001, respectively, for time-limited regression analysis of the change in endorsing warning labels immediately before and after policy change occurred, data not shown); in Canada, the rise in endorsing warning labels between 2010/2011 and 2013/2015 corresponds to the addition of new picture warnings and increase in warning size (AOR=1.35, p<0.001 for change in endorsement immediately before and after policy change occurred, data not shown); in Australia, endorsing price rose markedly immediately after the large tax increases in 2010 (AOR=1.77, p<0.001 for change in endorsement immediately before and after policy change occurred, data not shown) and also rose in the UK in 2010 when tobacco tax increases started to exceed inflation rates (AOR=1.56, p<0.001 for change in endorsement immediately before policy change until the end of the study period, data not shown). None of the other non-linearities can be readily linked to policy changes.
Reasons for thinking about quitting and making a quit attempt
Table 4 shows associations between endorsing reasons and making a quit attempt at follow-up wave. The group of ‘concern/risk’ reasons was most strongly associated with making a quit attempt, particularly ‘concern for personal health’, with those endorsing this reason having a nearly twofold greater odds of making a quit attempt without adjustment for other reasons, and a nearly 70% greater odds of making a quit attempt after adjusting for every other reason; the strength of this relationship was consistent across the four countries. The group of ‘assistance’ reasons and the group of ‘information’ reasons were each positively associated with making a quit attempt after adjusting for all reasons not included in the respective group.
Making a quit attempt as a function of endorsement of reasons that led smokers to think about quitting
Individual reasons positively associated with making a quit attempt after adjustment for all other reasons included advice from a health professional, advertisements about health risks, warning labels, setting an example for children, society disapproves and availability of helpline/quitline. The strength of these associations was generally consistent across the four countries, though there were exceptions for advice from a health professional and setting an example for children, both of which were more strongly associated with making a quit attempt in the USA than in the UK. Finally, there was a 9% greater odds of making a quit attempt for each increase of one in the total number of reasons endorsed.
Discussion
Results from this study show that between 2002 and 2015, smokers' concern for personal health was the most frequently endorsed reason for thinking about quitting in the UK, Canada, the USA and Australia, and across all reasons to quit smoking, concern for personal health had the strongest association with making a quit attempt at follow-up wave. These findings are consistent with numerous studies citing concern for one's health as the number 1 reason smokers report for wanting to quit,1 former smokers report for having quit11 and quit attempters report as their trigger for attempting to quit.12
The tobacco control policies that have been implemented in each of these countries during the study period (see table 1) provide some context for interpreting the between-country differences and within-country changes in smokers' reasons for thinking about quitting over time. We observed a non-linear trend for ‘concern/risk’ reasons in the UK such that endorsement of these reasons tended to rise until 2005/2006, declined for several years and then began to rise again after 2010/2011 (see online supplementary figure S1). These changes can likely be attributed to the dramatic changes in media expenditures in the UK during this time, which rose and peaked in 2004/2005, dropped to £0 in 2010/2011, and then started to rise again. We also observed overall upward trends in thinking about quitting due to smoking restrictions in the UK, consistent with the implementation of smoking bans in the UK during the course of our study period.
Several tobacco control policies were implemented in Canada during the course of our study period including bans on tobacco advertising, smoking in indoor public places and point-of-sale tobacco displays, and we found that smokers in Canada endorsed an average of 6.3 reasons out of the 12 reasons evaluated, with greater number of reasons endorsed being associated with greater odds of making a quit attempt. Canada introduced a revised set of pictorial health warnings covering 75% of the front and back of cigarette packs in 2012, a policy that has been shown to be associated with increased motivation to quit,13–15 but we observed a steep increase in Canadian smokers' thinking about quitting due to warning labels after 2008, which precedes the introduction of the new labels, so not all of the increase can be attributed to the new warnings.
During the course of our study period, the USA created a national network of telephone helplines and raised the federal excise tax on cigarettes, and many states also raised cigarette taxes and adopted clean indoor air laws, all of which are positively associated with quitting thoughts/behaviors.16–20 Our data show that, indeed, there were overall upward trends in endorsement of ‘public/society’ and ‘assistance’ reasons during the course of the study period.
Interestingly, we observed a spike between 2008 and 2011 in the prevalence of US smokers reporting that warning labels on cigarette packs made them think about quitting. Although pictorial warnings were never added to cigarette packs in the USA, this corresponds to the time when the US Food and Drug Administration was granted regulatory authority over tobacco products and issued regulations to add pictorial health warnings to cigarette packs.21
For Australia, online supplementary figure S1L shows a steep increase in endorsing the price of cigarettes reason following the tobacco tax increase of 25% in 2010. Cigarette price/tax increases are a key component of tobacco control strategies, and these data demonstrate that the greatest increases in price are associated with the greatest increases in thoughts about quitting, but this motivator diminishes over time (eg, see online supplementary figure S1L, Canada), suggesting that large and repeated increases in price are needed to sustain their impact. In Australia, we also observed a steep increase in thinking about quitting due to cigarette pack warnings following the introduction of graphic health warnings to 30% of the front and 90% of the back of cigarette packs in 2006 and then another steep increase following the introduction of larger graphic health warnings (75% of the front and 90% of the back) of cigarette packs coupled with standardised or plain packaging requirements in late 2012.
The findings we report here are generally consistent with those reported by Hummel et al,2 who evaluated policy triggers for thinking about quitting in several European countries and found that price was the trigger most commonly endorsed across countries. Price was also the policy-specific reason that we found to be most commonly endorsed in the UK, Canada, the USA and Australia, but we also report similarly high endorsement of concern/risk reasons in all four countries (which were not assessed by Hummel et al). We also observed changes in some non-policy-specific reasons over the course of our study period including a noteworthy upward trend in thinking about quitting due to receipt of advice to quit from a health professional among smokers in the USA and Australia. Further, this reason was relatively strongly associated with making a quit attempt, particularly in the USA, demonstrating the importance of health professionals ensuring they give smokers appropriate advice.
Early data from the ITC-4C (2002–2005) showed that smokers in the UK were less likely to attempt to quit than smokers in Canada, the USA or Australia,22 which seems to line up with the current finding that smokers in the UK are less likely than their counterparts in the other three countries to endorse concern for personal health as a reason for thinking about quitting, which is the reason most strongly associated with making a quit attempt. However, further research is needed to determine whether there are differences among these countries in cognitive or behavioural factors such as quitting self-efficacy and nicotine dependence, and how such differences may relate to quit attempt rates in each country. Additional limitations of the current study include relatively low survey response rates (though prior analyses have shown that our participant characteristics correspond well to characteristics of responders to national benchmark surveys), and an average attrition rate of ∼30% (though we adjusted our analyses for characteristics known to vary with respect to retention). Balanced against our study limitations is our use of a large nationally representative sample of smokers from four countries who were surveyed over the course of more than a decade, our use of GEEs, which accounted for repeated analyses of the same respondents at different points in time, and our ability to examine changes both between and within countries.
Conclusions
Smokers' concern for personal health is the most frequently endorsed reason for thinking about quitting in the UK, Canada, the USA and Australia, and it is strongly associated with making a quit attempt. Various changes in endorsement of other reasons over the course of the study period were observed in these countries, some of which are likely linked to policy changes. Also, increases in the number of reasons endorsed was associated with increases in the likelihood of making a quit attempt, which suggests a benefit for tobacco control efforts that broaden the number of reasons smokers think about quitting.
What this paper adds
Cigarette smokers commonly cite health concerns, social concerns and cost as reasons for wanting to quit smoking.
Tobacco control policies have been implemented to increase quit rates in the UK, Canada, the USA and Australia.
Absent from the literature is a comprehensive multinational evaluation of smokers' reasons for thinking about quitting situated within the context of the tobacco control policies that were implemented to directly or indirectly increase quitting.
Our findings show that between 2002 and 2015, smokers' concern for personal health was consistently the most frequently endorsed reason for thinking about quitting in each country and across waves, and was most strongly associated with making a quit attempt.
Endorsement of health warnings, and perhaps price (particularly in Australia), appears to peak in the year or so after the change is introduced, whereas other responses were not immediately linked to policy changes.
References
Footnotes
Contributors KAK conducted the statistical analysis and drafted the initial manuscript; all authors contributed to study conceptualisation, data interpretation and manuscript revision.
Funding The ITC Four Country Survey has been funded by the US National Cancer Institute (P50 CA111326, P01 CA138389, R01 CA100362, R01 CA090955), Canadian Institutes of Health Research (57897, 79551 and 115016), Commonwealth Department of Health and Aging, National Health and Medical Research Council of Australia (265903, 450110, 1005922 and 1106451), Cancer Research UK (C312/A3726, C312/A6465 and C312/A11039, C312/A11943), Robert Wood Johnson Foundation (045734) and Canadian Tobacco Control Research Initiative (014578). Additional support was provided to Geoffrey T Fong from a Senior Investigator Award from the Ontario Institute for Cancer Research and a Prevention Scientist Award from the Canadian Cancer Society Research Institute.
Disclaimer None of the sponsors played any direct role in the design or conduct of the study, the collection, management, analysis or interpretation of the data, the preparation of the manuscript, or the decision to submit the manuscript for publication.
Competing interests KMC has received grant funding from Pfizer to study the impact of a hospital-based tobacco cessation intervention and also has served as an expert witness in litigation filed against the tobacco industry.
Ethics approval The ITC-4C study protocol was approved by the institutional review boards/research ethics boards of the University of Waterloo (Canada), Roswell Park Cancer Institute (USA), University of Strathclyde (UK), University of Stirling (UK), The Open University (UK), and The Cancer Council Victoria (Australia).
Provenance and peer review Not commissioned; externally peer reviewed.