Future nicotine use preferences of current cigarette smokers: Findings from the 2020 International Tobacco Control Four Country Smoking and Vaping Survey

Abstract

Introduction:

Consumer preferences should be important factors that are considered when developing health policies and interventions. This paper examines the prevalence of, and factors associated with, consumer preferences regarding smoking behaviour one to two years in the future.

Methods:

At least weekly cigarette smokers in the 2020 wave of the International Tobacco Control Four Country Smoking and Vaping Survey (USA, Canada, England and Australia) (N=8642) were asked if they preferred to continue to smoke or to quit with or without an alternative nicotine product (ANP) over the next 1–2 years.

Results:

Country-specific weighted data showed 21.5% preferred to continue smoking and 8.0% were uncertain, leaving 70.6% preferring to quit: 13.7% using an ANP and 56.9% completely quitting nicotine. Apart from interest in quitting, the main predictors of preferring to quit were history of vaping, being aged 55 and over, smoking weekly, worrying about smoking harms, regretting starting and believing vaping is less harmful relative to smoking. Among those preferring to quit, preferring to use ANPs in future was very strongly associated with current vaping (especially daily), being younger, living in England, reporting strong urges to smoke, believing vaping is much less harmful than smoking, and not strongly regretting starting to smoke, and not wanting to quit.

Discussion and Conclusions:

A significant minority of smokers preferred not to quit, at least in the next year or two. Both interest in quitting and preference for ANPs over complete cessation were associated with similar covariates, including interest in vaping.

INTRODUCTION

Most smokers in western countries regret ever starting smoking (85–95%) yet continue to smoke [1,2]. Most smokers make numerous failed attempts before they permanently quit smoking, if they ever do [3,4]. In countries with advanced tobacco control policies, 30–40% of smokers report making a quit attempt every year, with an average of 1–2 quit attempts per year [3]. There appears to be some ambivalence and/or resistance to quitting. In one recent study of smokers from the same study reported on here only 35% wanted to quit “a lot” and over the next two years only 64% of this group reported a quit attempt [5]. In keeping with this, only around 30% reported plans to quit within the next six months, suggesting a lack of urgency [5]. At the other extreme, this study also identified 12% who did not want to quit and 35% with no plans to quit, suggesting that some are content to continue smoking [5]. However, this inference is based on past behaviour and more immediate plans and desires, which are likely influenced by perceptions of their ability to quit, and as far as we are aware, no one has looked at what smokers’ medium/long-term preferences are.

The choices people make in any given behaviour are shaped by their perception of available options and the context in which the choice is made [6]. For example, what people might want in an ideal situation may differ markedly from what they choose in their current context, or the context they assume or think is likely to be present. One important contextual factor is the availability of alternatives, in this case, for smoking cessation, the possibility of substituting an alternative nicotine product (ANP). There is an increasing range of ANPs available [7], although this varies considerably by jurisdiction, including between the four countries included in the International Tobacco Control Four Country Smoking and Vaping Survey (ITC 4CV survey) [5,8]. Although the long-term effects of use of ANPs is not well established for most (except low toxin oral snuff as used in Scandinavia) [9], what evidence there is points to likely marked reductions in risks for some of these products and arguably all [10,11]. It is thus a reasonable assumption for an individual who smokes to make that switching to one of these products is likely to reduce their current health risks. In recent years, the proportion of smokers and ex-smokers using ANPs has increased [12]. Most ANP use is in the form of nicotine vaping products (NVP), but heated tobacco products and/or various forms of oral tobacco or oral nicotine products have also become more prevalent in some countries, such as England and Canada [8]. There is increasing evidence that much of this due to smokers exploring alternatives to smoking [13], and the evidence strongly suggests that ANP does not support continued smoking [14,15]. To date, the evidence shows that ANPs can facilitate quitting [16,17], and their use is associated with increased interest in smoking cessation [16–18], and continued use carries a small fraction of the health risks of smoking [13]. However, of the smokers who want to quit smoking, it is unclear what proportion expect or prefer to use an ANP rather than to be abstinent from nicotine. This is important in the context of increased discussion around options to eliminate the widespread use of cigarettes [19].

This paper examines factors that are associated with cigarette smokers’ preference for their smoking one to two years in the future (medium term preference). We explore the associations of a range of motivational variables and vaping-related measures with the key choices of continuing to smoke cigarettes (vs not) and among those preferring to quit smoking, of their choice between complete abstinence and use of ANPs.

METHODS

Data source and participants

Data came from the ITC Four Country Smoking and Vaping Wave 3 (ITC-4CV3) Survey conducted in Australia, Canada, England and the USA between February-June 2020. The ITC-4CV Survey was an online cohort study conducted in the above four countries. Respondents were adults aged ≥18 years, and were recruited by survey firms in these countries (more information can be found at https://itcproject.org/methods/technical-reports/). There were 8642 respondents (Australia: n=1213; Canada: n=2633; England: n=3057; US: n=1739) who currently smoked cigarettes at least weekly and were asked about the key question of interest (i.e., their future smoking preferences). More details about the ITC 4CV Surveys have been reported elsewhere [20,21].

Measures

Preferred nicotine use for the next 1–2 years (medium-term)

Current cigarette smokers were asked “Which of the following would you MOST prefer to do over the next 1–2 years”: 1. Continue to smoke cigarettes; 2. Quit smoking cigarettes, by switching completely to an alternative nicotine product (e.g., vaping product, heated tobacco product); 3. Quit smoking cigarettes and not switch to another nicotine product; and 4. Don’t know/refused. The time period chosen was to ground the choice in an assumed reality but be distant enough to be able to ignore short term contingencies.

Smoking and vaping related variables

Current smokers were those reporting daily cigarette smoking or weekly smoking. Smokers were asked if they considered themselves addicted to cigarettes and strength of urge to smoke in the last 24 hours (options were: did not feel the urge to smoke, slight, moderate, strong, very strong, extremely strong). Smokers were also asked whether they had made quit attempts (yes vs. no) in the previous period (since last survey wave or in the past two years). Self-efficacy for quitting was assessed by asking: “If you decided to give up smoking completely in the next 6 months, how sure are you that you would succeed?” (“not at all sure”, “slightly sure”, “moderately sure”, “very sure”, “extremely sure” and “don’t know”).

Vaping status was coded as follows: “never vaper” (which included those who reported no more than a puff or two on a vaping device), “ex-vaper” (those who had used more than once or twice but no longer used at least monthly), “current non-daily vaper” (current weekly or monthly use) and “current daily vaper”.

Motivational measures

We used a number of measures routinely collected in ITC surveys [1,5,22]. Respondents were asked about their plans and motivation (wanting) to quit via the following questions: “Are you planning to quit smoking?” (“within the next month”, “within the next 6 months”, “sometime in the future, beyond 6 months”, “not planning to quit” and “don’t know”); and “How much do you want to quit smoking?” (not at all, a little, somewhat, a lot, don’t know). Concern about the harms of smoking were assessed by asking smokers the following questions: “To what extent, if at all, has smoking cigarettes damaged your health?” (not at all, just a little, a fair amount, a great deal, don’t know); and “How worried are you that smoking cigarettes WILL damage your health in the future?” (not at all worried, a little worried, moderately worried, very worried, don’t know). Regret for having started smoking was assessed by asking “If you had to do it over again, you would not have started smoking cigarettes” (strongly agree, agree, neither agree nor disagree/don’t know, disagree/strongly disagree).

Belief about the relative harm of vaping was assessed by asking “Compared to smoking cigarettes, how harmful do you think vaping is?” (much less harmful than smoking cigarettes, somewhat less harmful than smoking cigarettes, equally harmful to smoking cigarettes, somewhat more harmful than smoking cigarettes, much more harmful than smoking cigarettes, don’t know).

Sociodemographic variables

In addition to country (USA, Canada, England, Australia), other sociodemographic measures used were gender (male, female) and age (18–24, 25–39, 40–54, 55 years and older; and “<=39” vs. “>=40” in some analyses). Due to differences in income and educational systems across countries, only relative levels of income and education were used. “Low” level of education referred to those who completed high school or less in Canada, the US and Australia, or secondary/vocational or less in England; “moderate” meant community college/trade/technical school/some university (no degree) in Canada and the US, college/university (no degree) in England, or technical/trade/some university (no degree) in Australia; and “high” referred to those who completed university or postgraduate studies in all countries. Household income was also categorised into three levels (“low”, “medium” and “higher”), with the tertiles roughly comparable across the four countries; and those who did not provide income information were included in the “not reported” group.

Data analysis

Descriptive statistics on prevalence data were reported using data weighted to each country, but adjusted analyses which include the weighting variables were conducted on unweighted data. Descriptive statistics (chi square tests) were used to examine sample characteristics by future nicotine use preferences. Among those who preferred to quit, logistic regressions were employed to identify independent predictors of a preference to quit smoking and preference for vaping over complete abstinence. In initial exploration of predictors of a preference for quitting smoking, the preliminary analyses indicated that those who did not know what they wanted to do responded in a similar pattern as those who preferred to “continue to smoke cigarettes”. As a result, these two groups were combined for the binary comparison. We considered this categorisation as conceptually sensible because not knowing can reasonably be thought of as being more likely to continue their current behaviour. We also performed sensitivity analysis by restricting the analyses to current daily smokers and by excluding subsets of variables to explore possible paths of influence. This included adding the variables “want to quit” and “plan to quit” to the analysis of quit preference, having initially excluded them because of their strong interrelationships with a preference to quit. In all analyses, a p value <0.05 was considered statistically significant. All analyses were conducted using Stata Version 16.0 [23].

RESULTS

Sample characteristics

Table 1 shows the characteristics of the overall sample of daily and weekly smokers (n=8642) stratified by nicotine use preferences 1–2 years in the future. Overall, with data weighted to population estimates for each country, 21.5% of the sample preferred to continue smoking in 1–2 years’ time and 8.0% did not know. Most preferred to quit (70.6%), with the majority of these preferring to quit all nicotine use (56.9%) and a minority preferring to switch to an ANP (13.7%). Bivariate analyses show that females and weekly smokers were less likely to choose continuing to smoke, compared to males and daily smokers. As might be expected, those having no current plans or desire (i.e., not wanting) to quit were more likely to prefer continuing to smoke (47% with no plan, and 73% with no desire) compared to those who had plans and desires. When treated as a binary (Prefer to quit vs not), the relationships are large (Want to quit χ²(3) = 2103.1, p ≤0.0001; Plan to quit χ²(3) = 195.7, p <0.0001).

Table 1.

Sample characteristics, by future nicotine use ambitions (weighted estimates)

	% of category weighted (unweighted) (n=8642)	Continue to smoke (n#=1799), %	Quit by switching to ANPs (n=1847), %	Quit without switching (n=4402), %	Don’t know (n=594), %
Total	100	21.4	13.7	56.9	8.0
Country		p <0.001~
Canada	30.9 (30.5)	15.6	11.4	65.8	7.2
USA	20.2 (20.1)	22.5	8.6	58.9	10.1
England	35.4 (35.4)	26.3	19.1	47.2	7.4
Australia	13.5 (14.0)	20.6	12.2	59.0	8.3
Smoking status		p <0.001
Weekly	11.4 (15.5)	14.8	19.5	57.4	8.3
Daily	88.6 (84.6)	22.3	12.9	56.8	7.9
Sex		p <0.001
Male	53.2 (50.4)	24.3	14.5	54.0	7.1
Female	46.8 (49.7)	18.3	12.7	60.1	8.9
Age, years		p <0.001
18–24	8.3 (20.5)	20.6	24.1	49.5	5.9
25–39	32.3 (20.0)	20.8	16.7	54.5	7.9
40–54	27.3 (24.9)	21.3	12.7	57.2	8.8
55+	32.2 (34.6)	22.5	8.8	60.9	7.8
Wanting to quit		p <0.001
No desire	14.0 (13.8)	73.2	5.2	9.5	12.1
A little	21.0 (21.2)	32.3	14.6	41.3	11.8
Somewhat	32.0 (32.0)	11.2	16.4	64.0	8.4
A lot	33.1 (32.9)	2.8	14.1	79.8	3.4
Planning to quit		p <0.001
No current plan	34.3 (33.8)	47.2	8.6	29.7	14.5
Beyond 6 months	31.6 (31.5)	13.1	15.5	64.7	6.7
Between 1–6 months	23.0 (23.6)	3.4	17.9	76.8	1.9
<1 month	11.2 (11.2)	3.2	15.5	77.5	3.8

^#In some analyses the numbers were less than the total due to missing cases.

^~p values in this column are based on chi-square analyses.

ANP, alternative nicotine products.

Factors associated with preference for quitting smoking

Table 2 presents an analysis of differences between those who prefer to stop smoking (either with or without using an ANP) compared to those who prefer to continue smoking. We report results from both bivariate analyses and those from a multivariate logistic regression where all listed variables in the table were included. We did not include either wanting to quit or quit plans in the analysis of basic interest in quitting as they are very highly correlated, but included them in the analysis of preferred outcome for those preferring to quit.

Table 2.

Factors associated with preference for quitting smoking (n=8642), and use of alternative nicotine (among smokers expecting to quit, n=6249)

Factors	n 8642	% preferring to quit smoking	Full model, for preferring to quit# aOR [95% CI] N=8399	P	n 6249	% use alternative nicotine (29.6%, n=1847)	Full model, for preference for alternative nicotine, n=6146 aOR [95% CI]	P
Age		p <0.001~				p <0.001
18–24	1770	77.2	0.89 [0.74, 1.08]	0.243	1367	44.9	1.78 [1.43, 2.21]	<0.001
25–39	1730	71.2	0.64 [0.55, 0.76]	<0.001	1231	39.0	1.67 [1.36, 2.06]	<0.001
40–54	2154	71.7	0.87 [0.75, 1.01]	0.066	1545	26.0	1.34 [1.10, 1.62]	0.004
55+	2988	70.5	Ref		2106	16.7	Ref	.
Gender		p <0.001				p <0.001
Male	4351	70.6	Ref		3071	34.6	Ref	.
Female	4291	74.1	1.16 [1.03, 1.29]	0.012	3178	24.7	0.84 [0.73, 0.96]	0.013
Country		p <0.001				p <0.001
Canada	2633	77.7	Ref		2046	26.4	Ref
USA	1739	69.6	0.69 [0.59, 0.81]	<0.001	1210	22.0	0.89 [0.73, 1.09]	0.268
England	3057	68.9	0.73 [0.63, 0.84]	<0.001	2107	41.2	1.55 [1.31, 1.83]	<0.001
Australia	1213	73.0	1.07 [0.88, 1.29]	0.500	886	19.5	1.17 [0.91, 1.48]	0.217
Education		p=0.013				p <0.001
Low	2158	69.9	Ref		1508	23.5	Ref	.
Moderate	4065	73.3	1.09 [0.95, 1.26]	0.207	2979	28.3	0.91 [0.76, 1.09]	0.331
High	2419	72.8	1.01 [0.86, 1.18]	0.948	1762	36.8	1.06 [0.87, 1.30]	0.568
Income		p <0.001				p <0.001
Low	2369	70.2	Ref		1663	24.8	Ref	.
Moderate	2470	72.3	1.03 [0.89, 1.19]	0.671	1785	30.2	1.08 [0.90, 1.30]	0.408
High	3347	74.7	1.04 [0.90, 1.20]	0.594	2499	33.6	1.10 [0.92, 1.32]	0.288
Not reported	456	66.2	0.95 [0.73, 1.24]	0.712	302	18.9	0.72 [0.49, 1.04]	0.080
Smoking status		p <0.001				p <0.001
Daily	7307	71.0	0.70 [0.59, 0.84]	<0.001	5191	26.8	0.81 [0.67, 0.97]	0.026
Weekly	1335	79.3	Ref		1058	43.0	Ref
Urges to smoke in last 24 hours		p <0.001				p=0.005
None	457	65.9	Ref	.	301	29.6	Ref	.
Slight	1454	71.4	1.32 [1.01, 1.73]	0.040	1038	32.0	1.39 [0.99, 1.97]	0.059
Moderate	2968	72.8	1.42 [1.09, 1.84]	0.008	2162	31.5	1.75 [1.25, 2.45]	0.001
Strong	1994	75.6	1.54 [1.17, 2.03]	0.002	1507	27.3	1.84 [1.29, 2.63]	0.001
Very Strong	931	74.0	1.28 [0.95, 1.74]	0.105	689	28.2	2.09 [1.42, 3.08]	<0.001
Extremely strong	756	69.7	1.14 [0.84, 1.56]	0.405	527	25.1	1.63 [1.08, 2.46]	0.021
Vaping status		p <0.001				p <0.001
Never vaper	3149	64.9	Ref	.	2045	9.5	Ref	.
Ex-vaper	2718	76.9	1.52 [1.32, 1.75]	<0.001	2013	18.1	1.72 [1.40, 2.11]	<0.001
Current non-daily	1535	76.7	1.39 [1.17, 1.67]	<0.001	1177	53.3	7.21 [5.80, 8.95]	<0.001
Current daily	1313	75.9	1.35 [1.12, 1.63]	0.002	996	66.0	12.33 [9.80, 15.51]	<0.001
Self-efficacy to quit		p <0.001				p <0.001
Not at all sure	3065	59.5	Ref	.	1825	25.5	Ref	.
Slightly sure	1943	79.7	2.50 [2.14, 2.91]	<0.001	1548	35.9	1.17 [0.96, 1.42]	0.113
Moderately sure	2277	83.1	3.34 [2.86, 3.90]	<0.001	1893	29.3	0.92 [0.76, 1.12]	0.399
Very sure	850	76.6	2.27 [1.84, 2.78]	<0.001	651	30.9	0.94 [0.73, 1.22]	0.654
Extremely sure	507	65.5	1.81 [1.42, 2.31]	<0.001	332	24.7	0.68 [0.47, 0.97]	0.034
Perceptions that smoking has damaged health		p <0.001				p <0.001
Not at all	1044	61.0	Ref		637	32.8	Ref	.
Just a little	2992	73.4	1.10 [0.92, 1.32]	0.301	2195	31.5	1.05 [0.83, 1.33]	0.677
A fair amount	2515	78.2	1.13 [0.93, 1.38]	0.208	1967	30.6	1.24 [0.97, 1.60]	0.092
A great deal	835	80.2	0.94 [0.72, 1.22]	0.636	179	26.7	1.20 [0.87, 1.65]	0.274
Don’t know	1210	62.2	1.06 [0.85, 1.31]	0.608	155	20.6	0.74 [0.54, 1.00]	0.049
Worries that smoking will damage health		p <0.001				p <0.001
Not at all worried	610	38.0	Ref		232	30.2	Ref	.
A little worried	2711	65.3	2.19 [1.77, 2.70]	<0.001	1769	31.9	1.55 [1.07, 2.26]	0.021
Moderately	2659	76.8	3.41 [2.73, 4.25]	<0.001	2041	32.7	1.61 [1.10, 2.35]	0.014
Very worried	2322	88.0	6.73 [5.25, 8.64]	<0.001	2044	24.0	1.28 [0.87, 1.88]	0.216
Don’t know	340	47.9	1.71 [1.23, 2.38]	<0.001	163	33.1	1.71 [0.97, 3.00]	0.062
Beliefs on how harmful vaping is, compared to smoking cigs		p <0.001				p <0.001
Much less	898	80.7	Ref		725	56.6	Ref	.
Somewhat less	2674	76.6	0.69 [0.54, 0.87]	0.002	2047	33.6	0.44 [0.36, 0.54]	<0.001
Equally harmful	2835	72.3	0.56 [0.44, 0.71]	<0.001	2049	21.5	0.28 [0.23, 0.35]	<0.001
Somewhat more	628	72.0	0.55 [0.41, 0.75]	<0.001	452	33.4	0.40 [0.29, 0.54]	<0.001
Much more	365	64.4	0.42 [0.30, 0.60]	<0.001	235	19.2	0.24 [0.16, 0.38]	<0.001
Don’t know	1141	60.1	0.63 [0.48, 0.83]	<0.001	686	14.0	0.31 [0.23, 0.43]	<0.001
Regret for having started smoking		p <0.001				p <0.001
Strongly agree	4450	82.5	3.26 [2.67, 3.96]	<0.001	3670	21.0	0.58 [0.44, 0.76]	<0.001
Agree	2180	69.2	1.72 [1.41, 2.10]	<0.001	1509	40.3	1.23 [0.93, 1.62]	0.150
No opinion	1302	52.6	0.95 [0.76, 1.17]	0.609	685	44.2	1.18 [0.87, 1.61]	0.284
Disagree/strongly disagree	692	54.2	Ref		375	42.9	Ref	.
Wanting to quit		p <0.001	[not included in the model]			p <0.001
No desire	1195	17.2			206	45.2	Ref	.
A little	1835	60.5			1110	42.0	0.67 [0.46, 0.98]	0.039
Somewhat	2768	81.5			2257	30.5	0.47 [0.32, 0.69]	<0.001
A lot	2844	94.1			2676	22.4	0.34 [0.23, 0.51]	<0.001
Planning to quit		p <0.001	[not included]			p=0.924
No current plan	2780	39.9			1108	30.1	Ref	.
Beyond 6 months	2616	82.8			2165	29.5	0.92 [0.74, 1.13]	0.422
Between 1–6 months	2107	93.0			1959	29.1	0.88 [0.70, 1.11]	0.290
<1 month	1139	89.3			1017	30.1	0.83 [0.63, 1.08]	0.166
QA in previous period		p <0.001				p <0.001
Yes	4876	78.9	Ref		3846	31.9	Ref	.
No	3766	63.8	0.59 [0.53, 0.67]	<0.001	2403	25.9	1.08 [0.92, 1.27]	0.325

^#This model contains all the variables listed (except wanting and planning to quit for preferring to quit) . In adjusted analyses the sample size was less than the total due to missing cases in some variables. Those who answered ‘don’t know’ to preference question (n=594 unweighted) were excluded from this analysis.

^~p values in this column and in the ‘use alternative nicotine’ column are based on chi-square analyses.

CI, confidence interval; OR, odds ratio.

Overall, as shown in Table 2, many of the predictor variables had similar bivariate and multivariate relationships with preferences. Those showing a consistent positive relationship with preferring to quit were: being female; from Canada; smoking weekly (compared to daily); having vaped (compared to all who never vaped); worries that smoking will damage health; regret for starting smoking; believing that vaping is much less harmful than smoking cigarettes; having some self-efficacy to quit (especially at moderate levels); and having made quit attempts in the previous period.

The relationship between age and smoking preferences reversed after adjustment for covariates. Those aged 55 and over were less likely to prefer to quit smoking than smokers aged 18–24 in the bivariate analysis, but multivariate analysis indicates that they were most likely to prefer to quit smoking, significantly more so than the 25–39 year group (adjusted odds ratio for this group = 0.64, 95% confidence interval 0.55 – 0.76; p <0.001) (Table 2). Additional logistic regression analysis excluding sets of variables indicated that the age effect was reduced when smoking frequency (daily/weekly) was added to demographics, while adding vaping status led to the first significant reversal (see Table S1, Supporting Information) but the reverse effect became even stronger with all the other measures in Table 2 added.

The relationship between perceptions of health damage done by smoking and preference disappeared in the multivariate analyses.

Preferred non-smoking pattern

Among those preferring to quit smoking, there was a consistent pattern of being more likely to prefer having an alternative nicotine product among those who were: of younger age; male; English; weekly smoking; having some urges to smoke; more frequent vaping; believing that vaping is much less harmful than smoking cigarettes; and having no desire (i.e., not wanting) to quit. Of the above predictors the largest effect is for vaping status with both current daily (adjusted odds ratio 12.33, 95% confidence interval 9.80–15.51) and non-daily (adjusted odds ratio 7.21, 95% confidence interval 5.80–8.95) vapers much more likely to prefer using an alternative, compared with never vapers.

The following were associated with preferring to use an alternative nicotine product in bivariate analysis, but not in multivariate analysis: high education; high income; higher levels of self-efficacy to quit (but highest at slightly sure); no or a little concern about past health effects from smoking; and reporting recent quit attempts. Unlike for preference for smoking, there were no reversals of effects for preference for using an alternative smoking cessation method.

The associations between the key motivational variables and preferring to quit smoking cigarettes (and preferring to switch to use an alternative form of nicotine among those preferring to quit smoke) remained largely the same in multivariate models including only demographics and demographics plus smoking-related variables, and when the multivariate analyses were restricted only to current daily smokers (relevant results for daily smokers can be found in Table S2, Supporting Information).

DISCUSSION

This study found that only a small majority (57%) of smokers want to quit using all forms of nicotine in the medium term, leaving notable groups who would prefer to use some form of ANP (14%), continue to smoke (21%) or are uncertain (8%). Our findings are consistent with previous studies showing most smokers do not want to continue smoking [24], but the proportion not clearly wanting to quit all nicotine is higher than we anticipated. The minority without a preference to quit, most of whom expressed a preference for continuing to smoke, were unsurprisingly largely those also reporting not wanting to and/or not having current plans to quit, demonstrating construct validity for the measure.

Among those preferring to quit, having no plans to quit was not significantly associated with increased likelihood of preferring to switch to an ANP while not wanting to quit was. Wanting to quit is conceptually related to perceptions of gain and loss [25,26], and thus ANPs being a potential means of mitigating the losses associated with quitting, makes them potentially attractive to those who otherwise do not want to change.

Our findings confirm that there is a sizeable minority of smokers, who despite years of public education and the increased denormalization of smoking, appear committed to their smoking or are at least reluctant to give it up. Some may be in this position because they see quitting as beyond them. Consistent with this, those with the lowest levels of self-efficacy were notably less likely to prefer quitting, especially in comparison with moderate self-efficacy. Some of the desire to keep smoking is also likely to be a result of perceived benefits of smoking [27,28]. It is likely that some, but not all, of these perceived effects are real [29,30], so attempts to demonstrate falsity of beliefs is likely to have limited impact. Possible substitutes for smoking that replace some of the perceived functions may be more effective [31,32].

The size of the minority of smokers and some of the factors associated with lack of interest in quitting requires a rethink of the assumption that seems to be implicit in tobacco control thinking, that if sufficiently educated, and not distracted by tobacco industry promotion, all smokers who want to quit can be motivated enough to succeed with current supports [33,34].

It is notable that experience of vaping was associated with preferring to quit smoking, and the relationship was equally as strong among ex-vapers compared to current smokers, so it does not seem to relate to current vaping per se, but to any experience, or factors that lead to tying it. Related to this, believing that vaping is as harmful or worse than smoking was independently associated with preferring not to quit smoking. Taken together, vaping, especially if seen as a lower risk alternative to smoking, would seem to increase preferences for quitting smoking and is plausibly a mechanism for changing the minds of some who would otherwise be committed to continue smoking. A recent paper using ITC data is consistent with this [35], as vaping was associated with increased quitting among those who had no plans up to two years prior to quitting. Use of alternatives may be a behavioural indicator of heightened interest in quitting smoking, or indeed a route to smoking cessation that does not result in losing some of the valued aspects of smoking [34,36]. Further, the finding that interest in alternatives was greater in younger smokers may suggest that interest will grow as innovation often flows from the young [37].

The finding that those currently using NVPs are much more interested in quitting using alternatives suggests that some of this group are pursuing this strategy, albeit unsuccessful at this point. Exploration might be increased with appropriate public education, at least in part to reduce misbeliefs about the relative harmfulness of vaping as compared to smoking [38]. This may be the only viable path to convince many of those currently not wanting to quit, as it is unclear if more could be done to convince them of the potential harms given the long term, high quality, education campaigns to which many would have been exposed. This assumes that any smokers newly attracted to vaping would share the same interests in quitting as those who already have some experience. Consistent with the argument that use of vaping might be a conduit to increased interest in quitting [39], we found that preferring use of an ANP to complete cessation was associated with similar factors to interest in quitting overall, in particular increased exposure to vaping and more positive attitudes to it. However, having urges to smoke was also predictive of preferring to switch products, suggesting some smokers may expect an alternative form of nicotine to help reduce cravings while quitting smoking. This latter finding relates to the negative effects of quitting rather than positive effects of smoking. We think it likely that these two effects are interrelated. The results are consistent with a model that suggests improved public education about the likely much lower risk of non-combusted forms of nicotine may lead to a reduction in the proportion of smokers who would prefer to continue to smoke [40]. In this regard, public communications that lead to inaccurate risk perception that vaping is likely to be as, or more, harmful than smoking may harm smokers by discouraging quitting [41]. That said, it is important to balance consideration of the risks and benefits of NVPs and other ANPs [38,42]. We found nothing in our results to suggest any possible negative effect of use of ANPs in discouraging quitting. Our results indicate that among those preferring to quit, those who lived in England, where NVPs are more readily available and encouraged by health authorities ⁸, were more likely to prefer to use ANPs in future.

That all said, it remains the case that the majority of smokers, particularly among those preferring to quit, prefer to quit all forms of nicotine. Where this is a viable path, we should provide whatever assistance we can to help smokers achieve it. We do not know whether this number has declined with the availability of alternative products or whether it is attracting a new group of previously reluctant smokers. That this preference for complete cessation exists among current smokers, also highlights the difficulty of quitting, as one can infer that many of these smokers would have already quit if they could do so easily [43]. We should not underestimate the challenges smokers are likely to face if smoking is increasingly restricted. However, removing the environmental triggers to smoke and reducing convenient access to cigarettes through reducing retail availability may also support these smokers to act on and to achieve their goal [44,45], but failing that, ANPs could act as a means of enhancing the impact of such strategies, or act as a second-best fall-back if they have limited impact.

This study has limitations. The sampling is such that even with the weighting used, precise country-level estimates of the prevalence of the medium-term options should be treated as approximations rather than precise estimates. It is possible that inclusion of other variables in the adjusted analyses may have changed some associations, but we can think of none that would lead us to change the specific conclusions. Finally, the speculation about the results and possible implications should be treated as hypotheses worthy of future exploration not as conclusions.

CONCLUSION

In conclusion, while just over half of all smokers would like to have quit both smoking and nicotine completely within a year or two, a sizeable minority would not, instead preferring either to continue smoking or to use ANPs. Unless this changes, societies are unlikely to effectively eliminate smoking. A significant number of individuals desiring to completely quit nicotine are unlikely to achieve success using currently approved forms of support. Given the demonstrated efficacy of some ANPs, and that some (e.g., some forms of smokeless tobacco), if not most, are much less harmful than smoking, there is a need to go beyond a cessation-only approach to tobacco control if only for those who prefer to continue nicotine use or are unable to quit smoking without it. In the context of a notable demand for ANPs [46,47], this would appear to be sensible ethical public policy. From the perspective of consumer empowerment, such an approach should be at the forefront of considerations by those working in the interests of smokers and those at risk of becoming smokers. To progress, efforts are required to inform current smokers and potential ones about the reduced risks of ANPs. Additionally, research is needed to work out what is needed to make ANPs more acceptable to smokers who appear unable or unwilling to quit all forms of nicotine. Furthermore, research should focus on finding more effective ways of helping people quit all forms of nicotine when this is their preference.