Associations between smoking to relieve stress, motivation to stop and quit attempts across the social spectrum: A population survey in England.

Smoking prevalence in several high-income countries is steadily declining but remains persistently high in 'lower' socioeconomic position (SEP) groups, contributing to inequities in morbidity and mortality. Smoking to relieve stress is a commonly endorsed motive for continued smoking; however, it remains unclear whether smoking to relieve stress has a negative impact on motivation to stop and future quit attempts and if so, whether associations are moderated by SEP. This was an observational study with cross-sectional and prospective survey data from the nationally representative Smoking Toolkit Study in England. A total of 1,135 adult smokers were surveyed at baseline, with 153 (13.5%) respondents followed up at 12 months. Respondents provided information on demographic, social and smoking characteristics. A series of multivariable logistic regression analyses was conducted. Bayes Factors (BFs) were calculated to explore non-significant associations. Smoking to relieve stress was commonly endorsed by respondents from both 'lower' (43.2% [95% CI = 39.4%, 47.0%]) and 'higher' (40.5% [95% CI = 35.9%, 45.1%]) SEP groups (p = 0.39). Smoking to relieve stress was associated with high motivation to stop at baseline (ORadj = 1.48, 95% CI = 1.03-2.12, p = 0.035) but not significantly with the odds of making a quit attempt at a 12-month follow-up, although the magnitude and direction of the effect was similar to that observed for high motivation to stop (ORadj = 1.49, 95% CI = 0.69-3.20, p = 0.3). Data were insensitive to detect moderation effects of SEP (BF = 0.90 and BF = 1.65, respectively). Smoking to relieve stress is a commonly endorsed motive and is associated with high motivation to stop but not significantly with the odds of making a quit attempt in the next 12 months, although the magnitude and direction of the effect was similar for both outcomes. There was no clear evidence of moderation by SEP, although data were insensitive to distinguish the alternative from the null hypothesis.

Introduction

Tobacco smoking remains one of the leading causes of premature illness and death, with global smoking-related deaths approximating 8 million people per year in 2019 [1]. Smoking prevalence in a number of high-income countries, including England, has been steadily declining for the past decade but remains persistently high in ‘lower’ socioeconomic position (SEP) groups [2]. In England, ‘lower’ SEP is associated with increased morbidity and mortality, of which smoking is a significant contributor [3]. To help reduce observed health inequities between ‘higher’ and ‘lower’ SEP groups [4], supporting people to stop smoking–particularly those from priority SEP groups–is urgently needed. The development of effective policies and interventions requires the identification of key barriers to smoking cessation, such as beliefs or motives that hinder smoking cessation attempts.

Motivational factors, including desire or intention to stop, have been identified as key predictors of smoking cessation attempts [5]. Although longitudinal studies show that smoking cessation (as compared with continued smoking) is associated with reduced stress, anxiety and depression–with equal improvements to mental wellbeing observed in adults with and without a mental health diagnosis [6]–approximately 40% of smokers in England indicate that they smoke to cope with stress or anxiety [7]. However, it remains unclear whether smoking to relieve stress undermines motivation to stop and subsequent quit attempts. According to the PRIME (Plans, Responses, Impulses, Motives, Evaluations) Theory of Motivation [8], plans to stop smoking influence behaviour only if they generate sufficiently strong impulses at relevant moments in time. However, competing motives (including the belief that smoking helps to relieve stress) may act to undermine both plans and subsequent quit attempts by generating strong, opposing impulses. It is therefore plausible that the belief that smoking relieves stress may lead to reduced motivation to stop or it may undermine intentions to stop by generating strong, competing impulses which contribute to continued smoking. Previous research has examined whether the belief that one is addicted to smoking (‘perceived addiction’) undermines motivation to stop and future quit attempts. However, contrary to predictions, perceived addiction to smoking was recently found to be positively associated with motivation to stop [9] and future quit attempts [10].

SEP refers to social and economic factors that influence the position that individuals inhabit within the structures of society, and which in turn influence their health [11]. These factors typically include (but are not limited to) income, educational attainment, home ownership/tenure, occupation and employment status. The term ‘lower’ SEP is used in this study to reflect the relatively average lower income and status of those in routine and manual occupational trades and those who are unemployed. Smokers from lower SEP groups face multiple environmental and social stressors, including financial and occupational hardship and instability, as well as poor physical and mental health [11]. They also face smoking-specific stressors such as a lack of social support and higher nicotine dependence (compared with smokers from more advantaged groups) [12, 13], which means that negative affect from withdrawal is likely experienced as more intense and stressful, also highlighting the role of negative affect control through continued smoking [14].

Several qualitative studies have highlighted the role of smoking to cope with stress in particular subgroups of smokers, including adults experiencing homelessness [15], smokers living with psychiatric conditions [16], heavy smokers [17] and older smokers [18]. However, it remains unclear whether smoking to relieve stress is more commonly endorsed in smokers from lower (compared with higher) SEP groups, but it is plausible given the frequent occurrence of high stress levels in lower SEP groups, coupled with the widely held belief that smoking relieves stress. In addition, the belief that smoking is beneficial for stress is likely weighed against perceived benefits of quitting, such as improved health and financial savings (which could help alleviate financial hardship) [19]. It is therefore important to examine not only whether smoking to relieve stress has a negative impact on motivation to stop and subsequent attempts to quit, but also whether these associations are moderated by SEP. Recent research indicates that although motivation to stop appears similar by housing tenure–a reliable indicator of SEP and predictor of smoking status [20]–and prevalence of past-year quit attempts are higher in adults living in homes that belong to a housing association or are rented from local authority (compared with other housing), people from lower SEP groups are less likely to succeed in quitting [21].

We therefore aimed to examine, in the nationally representative Smoking Toolkit Study, the associations of smoking to relieve stress with motivation to stop (measured at baseline) and quit attempts at a 12-month follow-up in England, and whether these associations are moderated by SEP. Specifically, we aimed to address the following research questions:

Do smokers who indicate that they smoke to relieve stress have reduced odds of high (vs. low) motivation to stop smoking, without and with adjustment for age, sex, SEP, children in the household, cigarettes smoked per day and number of quit attempts in the past year?

Do smokers who indicate that they smoke to relieve stress have reduced odds of making a quit attempt at a 12-month follow-up, without and with adjustment for age, sex, SEP, children in the household, cigarettes smoked per day and number of quit attempts in the past year?

Does SEP moderate any associations between smoking to relieve stress and a) high motivation to stop and b) quit attempts at a 12-month follow-up, without and with adjustment for age, sex, SEP, children in the household, cigarettes per day and number of quit attempts in the past year?

Methods

Study design and setting

This was an observational study involving cross-sectional and prospective survey data. The STROBE guidelines were used in the design and reporting of this study [22].

The study is part of the ongoing Smoking Toolkit Study (STS), which involves monthly, face-to-face, computer-assisted household surveys of adults aged 16+ in England [23, 24]. The sample is a hybrid of a random probability and quota sample, which results in a sample that is representative of the adult population of smokers in England. Interviews are held with one household member in selected geographic output areas until quotas are fulfilled. The quotas are based on factors influencing the probability of being at home (i.e. working status, age and gender). This hybrid form of random probability and quota sampling is considered superior to conventional quota sampling. Here, the choice of households to approach is limited by the random allocation of small output areas and rather than being sent to specific households in advance, interviewers can choose which households within these small geographic areas are most likely to fulfil their quotas. Therefore, unlike random probability sampling, it is not appropriate to record the response rate in the STS. Data included in the present study were collected across four waves (139–142) from respondents surveyed between April and July 2018 (when smoking to relieve stress was included in the STS). Attempts to recontact consenting past-year smokers via telephone are made 12 months later. Follow-up respondents are not remunerated. Up to eight attempts are made to follow up each consenting respondent, with calls made at different times of day across weekdays and weekends.

Initial analyses were conducted as part of a student project. An extended study protocol and analysis plan were pre-registered on the Open Science Framework prior to conducting the analyses (https://osf.io/efbn2).

Inclusion criteria

Respondents were included in the analyses if they were aged 16+ years and smoked cigarettes at the time of the baseline survey. Respondents who indicated that they did not smoke cigarettes but did smoke tobacco of some kind were not included (~0.4%) as they do not complete equivalent measures of dependence (e.g. cigarettes per day). Respondents with missing data on any of the variables of interest were excluded from the analyses.

Ethics statement

Ethical approval was granted by UCL’s Research Ethics Committee (0498/001). Respondents provided verbal informed consent prior to each interview.

Measures

Outcome variables

The first outcome variable was motivation to stop, measured at baseline with the validated Motivation To Stop Scale [25, 26]. Respondents are asked: “Which of the following best describes you?” Response options include: 1) I don’t want to stop smoking; 2) I think I should stop smoking but don’t really want to; 3) I want to stop smoking but haven’t thought about when; 4) I REALLY want to stop smoking, but I don’t know when I will; 5) I want to stop smoking and hope to soon; 6) I REALLY want to stop smoking and intend to in the next 3 months; 7) I REALLY want to stop smoking and intend to in the next month. To aid interpretation, responses were dichotomised into low (response options 1–5) and high (response options 6–7) motivation to stop, as previously done in studies using this variable [27, 28].

The second outcome variable was quit attempts at a 12-month follow-up, measured with a single item: “How many serious attempts to stop smoking have you made in the last 12 months? By serious attempt I mean you decided that you would try to make sure you never smoked again.” Responses were coded 0 for respondents who reported no quit attempt and 1 for those who reported 1+ quit attempts.

Explanatory variable

The explanatory variable was smoking to relieve stress, measured at baseline with a single item asking respondents to select from a list the motives that they believe are important in keeping them smoking. Those endorsing the response option “It helps me cope with stress or anxiety” were coded as 1, the rest as 0.

Covariates

Covariates measured at baseline included: age (16–24 years, 25–34 years, 35–44 years, 45–54 years, 55–64 years, 65+ years); sex (male, female); SEP, operationalised as occupational social grade and measured with the British National Readership Survey’s Social Grade Classification Tool [29], which comprises the categories AB (higher and intermediate managerial, administrative and professional), C1 (supervisory, clerical and junior managerial, administrative and professional), C2 (skilled manual workers), D (semi-skilled and unskilled manual workers) and E (state pensioners, casual and lowest-grade workers, unemployed with state benefits), with responses grouped into ‘lower’ (C2DE) and ‘higher’ (ABC1) social grade; children in the household (no vs. yes); cigarettes per day (converted to daily consumption for non-daily smokers who reported the number of cigarettes smoked per week); and number of quit attempts in the past year (0, 1, 2, 3, 4+).

We also captured housing tenure as an alternative indicator of SEP [20]. Respondents whose homes belonged to a housing association or were rented from local authority were categorised as ‘lower’ SEP. All other responses (i.e. houses which were ‘bought on a mortgage’, ‘owned outright by household’, ‘rented from private landlord’, ‘other’) were categorised as ‘higher’ SEP.

Data analysis

Analyses were conducted in RStudio v.1.2.5033

Data were weighted with the rim (marginal) technique [30] to match the analytic sample to the proportions of the English population profile on the dimensions of age, social grade, region, housing tenure, ethnicity and working status within sex. We report descriptive statistics for the baseline and follow-up samples. Any differences between the baseline and follow-up samples were assessed with Chi-square or t-tests, as appropriate.

To address RQ1 and RQ2, unweighted logistic regression analyses were conducted, without and with adjustment for all covariates. To address RQ3, the analyses were repeated, including the two-way interaction between smoking to relieve stress and SEP. As a planned sensitivity analysis, the analyses were repeated with housing tenure used as the indicator of SEP [20].

Bayes Factors

In the event of non-significant interactions, we had planned to calculate Bayes Factors (using an online calculator; www.bayesfactor.info) to examine whether the non-significant associations could best be characterised as evidence of no association, evidence of an association, or whether the data were insensitive to detect an association. As there was limited evidence to inform what effect sizes to expect for the potential interactions with SEP, we set the expected effect sizes to OR = 0.7, informed by previous research examining the association between enjoyment of smoking and future quit attempts [31]. The alternative hypothesis was conservatively represented by a half-normal distribution. Bayes Factors (BFs) can be interpreted as follows: BFs >3 lend support to the alternative hypothesis (over the null), BFs <1/3 lend support to the null, and BFs in-between reflect data being insensitive to distinguish the two.

Results

A total of 1,203 cigarette smokers were surveyed between April and July 2018, of whom 1,135 (94.3%; unweighted) had complete data on all variables of interest. A total of 153 (13.5%; unweighted) respondents provided data at the 12-month follow-up (see Table 1). Compared with the baseline sample, those responding to the 12-month follow-up survey were older (p < 0.001), more likely to be from higher SEP groups (p < 0.001) and less likely to have children in the household (p = 0.02). At baseline, a total of 43.2% (95% CI = 39.4%, 47.0%) of respondents from lower SEP groups indicated that they smoke to relieve stress, compared with 40.5% (95% CI = 35.9%, 45.1%) of respondents from higher SEP groups (p = 0.39).

Table 1

Demographic, social and smoking characteristics of current cigarette smokers at baseline and at the 12-month follow-up.

	Baseline sample		Follow-up sample
	Unweighted (N = 1,135)	Weighted (N = 1,170)	Unweighted (N = 153)	Weighted (N = 152)	p-value**
Sex, N (%)					0.78
Women	557 (49%)	550 (47%)	73 (48%)	69 (45%)
Men	578 (51%)	620 (53%)	80 (52%)	84 (55%)
Age, N (%)					<0.001
16–24 years	189 (17%)	194 (17%)	13 (8.5%)	13 (8.2%)
25–34 years	242 (21%)	287 (25%)	20 (13%)	24 (16%)
35–44 years	191 (17%)	224 (19%)	27 (18%)	34 (22%)
45–54 years	204 (18%)	206 (18%)	30 (20%)	29 (19%)
55–64 years	152 (13%)	133 (11%)	27 (18%)	24 (15%)
65+ years	157 (14%)	125 (11%)	36 (24%)	29 (19%)
SEP * , N (%)					<0.001
Higher	457 (40%)	455 (39%)	84 (55%)	82 (54%)
Lower	678 (60%)	715 (61%)	69 (45%)	71 (46%)
Housing tenure, N (%)					0.44
Higher	767 (68%)	905 (77%)	108 (71%)	121 (79%)
Lower	368 (32%)	264 (23%)	45 (29%)	31 (21%)
Children in the household, N (%)					0.02
No	760 (67%)	774 (66%)	116 (76%)	114 (75%)
Yes	375 (33%)	396 (34%)	37 (24%)	38 (25%)
Cigarettes per day, M (SD)	10.6 (8.0)	10.3 (7.8)	11.4 (8.1)	10.8 (7.7)	0.25
Number of quit attempts in the past year, N (%)					0.17
0	827 (73%)	859 (73%)	119 (78%)	119 (78%)
1	182 (16%)	184 (16%)	23 (15%)	21 (14%)
2	68 (6.0%)	74 (6.3%)	9 (5.9%)	10 (6.6%)
3	33 (2.9%)	31 (2.6%)	0 (0%)	0 (0%)
4+	25 (2.2%)	22 (1.9%)	2 (1.3%)	2 (1.1%)
Smoking to relieve stress, N (%)					0.58
No	657 (58%)	686 (59%)	85 (56%)	87 (57%)
Yes	478 (42%)	484 (41%)	68 (44%)	65 (43%)
Motivation to stop smoking, N (%)					0.32
Low	968 (85%)	1,004 (86%)	135 (88%)	136 (89%)
High	167 (15%)	166 (14%)	18 (12%)	17 (11%)
Quit attempt at the 12-month follow-up, N (%)
No	-	-	89 (58%)	88 (58%)	-
Yes	-	-	64 (42%)	64 (42%)

* SEP = socioeconomic position

** p-values pertain to comparisons of the baseline and follow-up samples (unweighted), with values <0.05 highlighted in bold

In unadjusted and adjusted analyses without the interaction, smoking to relieve stress was associated with 48% greater odds of high motivation to stop (ORadj = 1.48, 95% CI = 1.03–2.12, p = 0.035; see Table 2). The association between smoking to relieve stress and quit attempts at the 12-month follow-up was not significant (ORadj = 1.49, 95% CI = 0.69–3.20, p = 0.3). The adjusted associations between the exposure and outcome variables were unchanged in the sensitivity analyses with housing tenure (see S1 Table). In an unplanned sensitivity analysis with cigarettes per day coded as a categorical variable, the associations were similar (see S2 Table).

Table 2

Unadjusted and adjusted odds ratios for the associations of smoking to relieve stress with i) high motivation to stop and ii) quit attempts at the 12-month follow-up.

	High motivation to stop (N = 1,135)			Quit attempts at the 12-month follow-up (N = 153)
	OR^	95% CI	p-value	OR	95% CI	p-value
Smoking to relieve stress (ref = No)
Yes	1.75	1.26, 2.43	<0.001	1.64	0.86, 3.16	0.13
Sex (ref = Women)
Men	0.80	0.57, 1.11	0.2	0.31	0.09, 0.87	0.033
Age (ref = 16–24 years)
25–34 years	1.16	0.71, 1.94	0.6	3.00	0.38, 62.8	0.4
35–44 years	1.06	0.62, 1.83	0.8	2.09	0.27, 43.3	0.5
45–54 years	0.84	0.49, 1.47	0.5	1.33	0.15, 28.5	0.8
55–64 years	0.64	0.33, 1.20	0.2	1.50	0.17, 32.1	0.7
65+ years	0.42	0.20, 0.83	0.016	1.09	0.13, 23.2	>0.9
SEP* (ref = Higher)
Lower	0.69	0.50, 0.97	0.030	0.57	0.19, 1.56	0.3
Children in the household (ref = No)
Yes	1.39	0.99, 1.95	0.055	0.16	0.01, 0.83	0.082
CPD**	0.96	0.93, 0.98	<0.001	0.93	0.86, 1.00	0.074
Number of quit attempts in the past year (ref = 0)
1	5.72	3.83, 8.55	<0.001	5.35	1.70, 16.5	0.003
2	6.50	3.68, 11.3	<0.001	3.49	0.47, 17.3	0.2
3	5.96	2.68, 12.6	<0.001	-	-	-
4+	6.71	2.74, 15.5	<0.001	-	-	-
	High motivation to stop (N = 1,135)			Future quit attempt (N = 153)
	ORadj	95% CI	p-value	ORadj	95% CI	p-value
Smoking to relieve stress (ref = No)
Yes	1.48	1.03, 2.12	0.035	1.49	0.69, 3.20	0.3
Sex (ref = Women)
Men	0.96	0.67, 1.38	0.8	1.32	0.62, 2.85	0.5
Age (ref = 16–24 years)
25–34 years	1.04	0.60, 1.82	0.9	0.61	0.12, 2.97	0.5
35–44 years	0.97	0.53, 1.77	>0.9	0.40	0.08, 1.80	0.2
45–54 years	1.01	0.55, 1.83	>0.9	0.50	0.11, 2.19	0.4
55–64 years	0.96	0.47, 1.89	0.9	0.21	0.04, 0.98	0.051
65+ years	0.54	0.25, 1.14	0.11	0.33	0.07, 1.42	0.14
SEP* (ref = Higher)
Lower	0.72	0.50, 1.04	0.084	1.30	0.63, 2.70	0.5
Children in the household (ref = No)
Yes	1.18	0.78, 1.78	0.4	0.93	0.37, 2.30	0.9
CPD **	0.96	0.94, 0.99	0.007	1.00	0.96, 1.05	0.8
Number of quit attempts in the past year (ref = 0)
1	5.29	3.50, 8.00	<0.001	6.09	2.20, 19.0	<0.001
2	6.08	3.40, 10.7	<0.001	5.63	1.20, 40.6	0.044
3	5.41	2.37, 11.8	<0.001	-	-	-
4+	7.13	2.83, 17.1	<0.001	3.23	0.11, 90.9	0.4

* SEP = socioeconomic position

** CPD = cigarettes per day

^ = Odds Ratio

ORadj = adjusted for sex, age, SEP, children in the household, CPD and number of quit attempts in the past year

- could not be estimated due to low cell counts

In unadjusted and adjusted analyses, the two-way interactive effects of smoking to relieve stress and SEP on high motivation to stop (ORadj = 1.66, 95% CI = 0.14–24.3, p = 0.7) and quit attempts at the 12-month follow-up (ORadj = 0.31, 95% CI = 0.1–1.4, p = 0.12) were not significant (see S3 and S4 Tables for the ORs stratified by SEP). Results remained non-significant in the planned sensitivity analyses with housing tenure (ORadj = 0.34, 95% CI = 0.03–4.31, p = 0.4 and ORadj = 0.72, 95% CI = 0.14–3.8, p = 0.7, respectively).

Bayes Factors (BFs)

The calculation of BFs indicated that the data on the two-way interactive effects of smoking to relieve stress and SEP on high motivation to stop and quit attempts at the 12-month follow-up were insensitive to detect an association (BF = 0.90 and BF = 1.65, respectively). As an unplanned sensitivity analysis, we also calculated a BF for the non-significant association between smoking to relieve stress and quit attempts at the 12-month follow-up using the same prior as the above analyses, which indicated that data were insensitive to detect an association (BF = 0.46).

Discussion

Principal findings

This study aimed to examine whether smoking to relieve stress is associated with high motivation to stop at baseline and quit attempts at a 12-month follow-up, and whether any associations are moderated by SEP. Contrary to expectations, we found that smoking to relieve stress was associated with high motivation to stop smoking, but not significantly with the odds of making a quit attempt at the 12-month follow-up. Although non-significant, the direction and magnitude of the association of smoking to relieve stress with future quit attempts were similar to those observed for the association with high motivation to stop. We did not observe significant moderation by SEP; however, the Bayes Factors indicated that data were insensitive to detect an association (as opposed to lending support to the null hypothesis of no association). For this to be known, other studies with larger samples need to be conducted to detect any potential moderation effects by SEP.

The finding that smoking to relieve stress is associated with high (as opposed to low) motivation to stop is consistent with previous research indicating that the belief that one is addicted to smoking is positively associated with motivation to stop [9]. There are several plausible explanations for this finding. First, people who smoke to relieve stress may evaluate their smoking experience as biological or instrumental (as opposed to pleasurable), which may be perceived as unpleasant or annoying, hence leading to greater motivation to stop. Second, those who smoke to relieve stress may find smoking enjoyable but at the same time, find it a source of dissonance to use smoking as a ‘crutch’ to cope with stress, hence leading to greater motivation to stop. Third, it is also possible that there is a common factor (e.g., depression) that gives rise both to the belief that smoking helps to relieve stress and high motivation to stop. Previous research has found that smokers with more severe (compared with less severe) levels of depression report lower quitting self-efficacy (which may be influenced by positive outcome expectations, such as the belief that smoking helps to cope with stress) but higher motivation to stop [32]. Fourth, as both smoking to relieve stress and motivation to stop were measured at the same time, there is potential for reverse causation, whereby those highly motivated to stop are also more likely to identify motives for smoking that some people may judge negatively (e.g. they may judge it unpleasant to feel that smoking is needed to relieve stress).

The finding that smokers from both higher and lower SEP groups endorse the belief that smoking helps to relieve stress may indicate that smokers are not only more/less likely to endorse this belief depending on their level of actual stressors (which may, on average, be higher in smokers from lower SEP groups). Instead, this may be related to people’s perceived ability to cope with stressors. Smokers across the social spectrum may perceive themselves to be better/worse equipped to cope with stress (irrespective of whether they actually are), thus influencing their beliefs about the role of smoking (e.g. positive outcome expectancies).

Strengths and limitations

This study was strengthened by examining associations in a nationally representative sample of smokers in England. To our knowledge, this was the first study to examine the associations of smoking to relieve stress, motivation to stop and future quit attempts in addition to differences in smoking to relieve stress by SEP. However, our study also had several limitations. First, the sample size for the prospective association of smoking to relieve stress and future quit attempts was small, which led to imprecision. However, this was pre-determined by the period for which the funding was available (i.e., the question about smoking motives–including smoking to relieve stress–was included only in a few STS waves), and we attempted to mitigate this by calculating Bayes Factors to elucidate whether the data provided evidence of no association or whether they were insensitive to detect an association. Second, the follow-up sample differed from the baseline sample on important characteristics, including SEP. This means that, had our findings pertaining to the follow-up sample been statistically significant, they may not be generalisable to the general population of smokers. In addition, if those who were more likely to be followed up (i.e. older smokers, men, those from higher SEP groups) were also more or less likely to endorse the belief that they smoke to relieve stress, this may have distorted any association with the outcome. Third, although the Smoking Toolkit Study is a nationally representative survey on key sociodemographic and smoking characteristics, it may not be representative with regards to psychological characteristics such as motivation to stop or the motive that smoking helps to relieve stress and it also may not reach smokers who are the most disadvantaged. Similar to the previous point about the follow-up sample, this may mean our findings are not generalisable to the general population of smokers. Fourth, the survey during this study did not collect data on perceived stress levels; this would be a useful variable to consider in future research. Fifth, smokers may not have recalled (and reported) all quit attempts made in the 12 months preceding the follow-up survey [33, 34]. Sixth, the analyses pertaining to motivation to stop were cross-sectional and therefore the direction of the association is unclear.

Implications for research, policy and practice

Future research to unpick the potential explanations for the positive association between smoking to relieve stress and high motivation to stop would benefit from deploying qualitative methods and/or Ecological Momentary Assessments in smokers’ daily lives, allowing a closer examination of the temporal dynamics of competing (or synergistic) motives, evaluations and impulses [8, 35]. Further understanding of the relationship between smoking to relieve stress and future quit attempts is also required in studies with larger sample sizes. As many quit attempts are, on average, needed before quitting smoking successfully [36], a positive relationship between smoking to relieve stress and making a quit attempt in the future would likely have a positive impact on public health.

The finding that smoking to relieve stress is a commonly endorsed motive in smokers from both lower and higher SEP groups indicates that public health interventions may benefit from tackling such misperceptions about smoking and stress relief, with evidence from a recent systematic review and meta-analysis of longitudinal studies indicating that smoking cessation (compared with continued smoking) is associated with reduced stress, anxiety and depression [6]. Although the finding that smoking to relieve stress is associated with high motivation to stop (and possibly the likelihood of making a quit attempt in the future) may be perceived as a positive finding that does not warrant intervention, it may be more ethically defensible to highlight to smokers the large body of work indicating that smoking cessation leads to positive mental health outcomes and use this as a motive for stopping, thus correcting common misperceptions. In addition, if smoking is normatively perceived as a stress reducer, then it may encourage people to take up smoking. This merits further investigation.

At the same time (although not measured in the present study), research shows that many mental health professionals believe that smoking helps to alleviate stress and therefore refrain from offering appropriate smoking cessation support to their clients [37]. This suggests that the tackling of common misperceptions also needs to occur among practitioners. Public health campaigns have often focused on the well-documented physical health aspects of smoking, with less attention paid to the mental health benefits of quitting, including reduced stress, anxiety and depression [6] but also increased levels of happiness [38, 39]. With increasing evidence for how to best counter misperceptions without inadvertently reinforcing them in the areas of vaccine hesitancy and climate change (e.g. normative feedback, framing messages in terms of gain) [40-42], interventions targeting healthcare professionals may benefit from drawing on this growing literature.

Finally, prior research shows that although smokers from lower SEP groups try to quit at a similar rate to smokers from higher SEP groups, they are less likely to succeed [43]. It is plausible that people from lower SEP groups endorse a confluence of motivational, cognitive and social factors for continued smoking (e.g. smoking to relieve stress and boredom, lower quitting self-efficacy, social aspects of smoking), which may interact to make it more difficult to quit [44]. Future research should examine the potential clustering of such factors and whether certain patterns are more pronounced in lower (compared with higher) SEP groups, as this is needed to help reduce smoking-related health inequalities.

Conclusion

Smoking to relieve stress is a commonly endorsed motive and is associated with high motivation to stop but not significantly with the odds of making a future quit attempt, although the magnitude and direction of the effect was similar for both outcomes. There was no clear evidence of moderation by SEP, although data were insensitive to distinguish the alternative from the null hypothesis.

Unadjusted and adjusted odds ratios for the associations of smoking to relieve stress with i) high motivation to stop and ii) quit attempts in the next 12 months in the planned sensitivity analysis with housing tenure.

(DOCX)

Click here for additional data file.

Unadjusted and adjusted odds ratios for the associations of smoking to relieve stress with i) high motivation to stop and ii) quit attempts in the next 12 months in the unplanned sensitivity analysis with cigarettes per day coded as a categorical variable.

(DOCX)

Click here for additional data file.

Unadjusted and adjusted odds ratios for the associations of smoking to relieve stress with i) high motivation to stop and ii) quit attempts in the next 12 months in the higher SEP group.

(DOCX)

Click here for additional data file.

Unadjusted and adjusted odds ratios for the associations of smoking to relieve stress with i) high motivation to stop and ii) quit attempts in the next 12 months in the lower SEP group.

(DOCX)

Click here for additional data file.

12 Jan 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Jesse T. Kaye, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified what type you obtained (for instance, written or verbal, and if verbal, how it was documented and witnessed). If your study included minors, state whether you obtained consent from parents or guardians. If the need for consent was waived by the ethics committee, please include this information.

Once you have amended this/these statement(s) in the Methods section of the manuscript, please add the same text to the “Ethics Statement” field of the submission form (via “Edit Submission”).

For additional information about PLOS ONE ethical requirements for human subjects research, please refer to http://journals.plos.org/plosone/s/submission-guidelines#loc-human-subjects-research.

3. Please include additional information regarding the survey or questionnaire used in the study and ensure that you have provided sufficient details that others could replicate the analyses. For instance, if you developed a questionnaire as part of this study and it is not under a copyright more restrictive than CC-BY, please include a copy, in both the original language and English, as Supporting Information. If the original language is written in non-Latin characters, for example Amharic, Chinese, or Korean, please use a file format that ensures these characters are visible.

4. Please state whether you validated the questionnaire prior to testing on study participants. Please provide details regarding the validation group within the methods section.

5. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

6. Thank you for stating the following in the Competing Interests section:

"OP, MT, and LK have no conflicts of interest to declare. SC has provided expert consultancy to providers of UK life insurance and the pharmaceutical industry on matters relating to smoking cessations aids. LS has received a research grant and honoraria for a talk and travel expenses from manufacturers of smoking cessation medications (Pfizer and Johnson & Johnson). JB has received unrestricted research funding from Pfizer to study smoking cessation."

We note that you received funding from a commercial source: 'Pfizer and Johnson & Johnson'

Please provide an amended Competing Interests Statement that explicitly states this commercial funder, along with any other relevant declarations relating to employment, consultancy, patents, products in development, marketed products, etc.

Within this Competing Interests Statement, please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your amended Competing Interests Statement within your cover letter. We will change the online submission form on your behalf.

7. Please note that in order to use the direct billing option the corresponding author must be affiliated with the chosen institute. Please either amend your manuscript to change the affiliation or corresponding author, or email us at plosone@plos.org with a request to remove this option.

8. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

9. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

Additional Editor Comments (if provided):

The authors report results of an interesting observational study examining the relations between self-reported stress-relief smoking motives and motivation to stop smoking and follow-up quit attempts. In particular, the authors are interested in whether SEP moderates these associations. These represent important scientific questions to inform movement toward achieving health equity in smoking treatment and intervention. The authors provide a detailed pre-registration on OSF in line with open science principles. Two expert reviewers provided their thoughtful critiques and recommendations below. I would encourage the authors to address their comments in detail, which I would consider a major revision.

The primary project seems well-designed, although additional methods details requested by both reviewers would enhance the manuscript and aid in interpretation of the data. One significant concern is the very small sample size at follow up and in particular that this sample may not be representative of the larger initial sample (e.g., especially on the SEP moderator variable of interest). As reviewer 1 notes, the limitations should not simply be listed but rather discuss how they constrain the interpretation of the results. For the small follow up sample size and representativeness, this should especially be discussed in greater detail and provide more cautious interpretation of this follow up data given these concerns (see reviewer 2). Furthermore, I would encourage the authors to be careful with their language in the discussion to how they interpret non-significant effects per reviewer 1’s recommendations. Additionally, the description of the results in the discussion should mirror the findings in the results section – it is not appropriate to imply group differences when the data analyses do not reveal significant differences (e.g., discussion first paragraph, last sentence).

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This is a well-designed study with interesting results for their well-justified research question. However, the paper would be greatly improved by the authors paying more attention to the interpretation of their results in the Discussion.

1. L33-42. Abstract. Revise the Results and Conclusion after addressing the suggestions below.

2. The introduction is clear and sets the seen well and justifies the research question well.

3. L188-190. Methods. It appears CPD has been treated as an interval variable rather than as a categorical variable, as for quit attempts in past year. This makes assumptions about the relationship between CPD and the outcomes which may not be justified, and a categorical variable may be more appropriate (and indeed they could consider using the combination measure of dependence Heaviness of Smoking index based on CPD and TTFC).

4. L198. It appears just the baseline was weighted to the population, not the cohort that was followed up. Please clarify, and explain why the cohort was not weighted if it was not.

5. L211-214. The description of the alternative SEP variable based on housing tenure would be better moved to paragraph describing covariates (ending L190).

6. L231-234. Results. This sentence only refers to p-values, not effect size or interval estimates. The authors should review how they report their results, especially results that are ‘not statistically significant’ in light of recommendations such as these and many more: Wasserstein RL, Lazar NA. The ASA statement on p‐values: context, process, and purpose. Am Stat 2016; 70: 129–133. Greenland S, McShane B. Scientists rise up against statistical significance. Nature 2019; 567: 305–307.

7. L244-248. The OR of the interaction term is not informative to readers. The authors could consider reporting separately the ORs for high and low SEP subgroups, perhaps in a supplementary table, to demonstrate the magnitude and direction of the effect modification (even though it is not statistically significant, and the Bayesian analysis suggested it was insensitive to detect an interaction effect).

8. L249-255. This paragraph would more logically sit before the paragraph above about interaction.

9. L251-255. Be more precise: start the sentence with ‘The adjusted association between the exposure and both outcome variables was unchanged…’ not ‘Results were materially …’. Drop the next two sentences as not germane, they are just about the adjusted association between an alternative covariate and the outcome.

10. Table 2. Consider just reporting an overarching p-value for the age and quit attempt categorical variables, rather than p-values for each category. Should also include the n for each analysis, and a footnote explaining what is adjusted for in the adjusted ORs – all variables. Consider also reporting the unadjusted association of the co-variates with the outcomes. Please check the journal guidelines if the precision of the reported p-values is correct: typically I would report 0.08 (not 0.084), and only go to 3 decimal points if less than 0.01.

11. Supplementary table – as above.

12. Discussion. Principal findings. Second and third sentence. Reconsider how you report non-significant findings. This does not mean that there is no association, just that you were unable to detect one. The point estimate of the association with making a quit attempt is of similar magnitude and direction as the association with motivation – however this is based on a smaller sample with less power. So don’t assume no impact, indeed consider discussing WHY there may also be an impact on quit attempts. The last sentence of this paragraph is inconsistent with the results – there is small non-significant difference between the groups: they are similar.

13. Discussion. Principal findings. Second paragraph. There is only scant discussion of possible reasons for the authors’ principal findings. The smoker’s belief that they are addicted is true, but the belief that they are motivated to stay smoking because they believe smoking helps them cope with stress is based on a misperception. The authors get close to explaining the link in the fourth sentence, but this needs to be stepped out more clearly. This should come before the previous sentence as that is about a methodological reason why this association may be spurious due to reverse causation. Might there also be a common factor that causes both the explanatory and outcome factor? The authors need to also provide some explanation for other principal findings, e.g. no difference in the belief that they are motivated to stay smoking because they believe smoking helps them cope with stress in the two SEP groups, before jumping to the need for more research, strengths and limitations, and implications for practice. These explanations need to better contextualise their new results with previous research.

14. Discussion. Strengths and limitations. Rather than just list these, the authors need to consider how they are likely to have altered their results and interpretation and implications of the results.

15. Discussion. Implications. First paragraph, second sentence. But if the perception is associated with increased motivation (and possibly increased likelihood of making a quit attempt in the future), this may not be a problem that needs to be corrected! Please discuss. In contrast, in the next sentence, this same misperception being held by health professionals definitely is a problem (which the authors have not measured or examined) and needs to be countered as it leads to less support. There is now increasing research evidence about how best counter such misperceptions without inadvertently reinforcing them, eg about vaccines and climate change, that could be referred to here.

16. Discussion. Implications. Second paragraph. The findings suggest that addressing misperceptions will be equity neutral, as similar association in high and low SEP. The paper does NOT address the association between stress and quitting, there are no implications from the results about addressing stress itself.

17. Discussion. Implications. Last paragraph. The first sentence is well substantiated, but the second sentence is not substantiated. Indeed, from other research, motivation seems mainly associated with initiating but not sustaining quit attempts.

18. Conclusion. The second sentence is correct and is a more correct description of the last clause of the first sentence.

Note that I am insufficiently familiar with Bayesian statistics to adequately review those methods.

Reviewer #2: The authors have presented an interesting manuscript that explored the relationship between using smoking to relieve stress and motivation to quit and future quit attempts in a UK sample of people 16 years and older. I think the authors did a good job laying out the premise for the analysis and I do think that there is merit in understanding how stress and coping with stress impacts smoking behaviors. The relationship is particularly salient among low SES populations as often we are inclined to advise people to quit but we don’t offer strategies to help them with the many stressors that lead them to smoke in the first place. I did have some questions about the methods and conclusions, which I will highlight below.

Methods

- Providing additional details on the Smoking Toolkit Study or a reference would be helpful. For example, I had the following questions:

o The sample was hybrid of probability and quota sample. Could you share how much was a probability based sample?

o Were respondents to the survey incentivized?

o How many waves are there for this survey?

o How were attempts made to reconnect with survey participants at 1 year follow-up? Why were there only 153 people at follow-up from such a large sample at baseline. Was the survey designed to be longitudinal? What steps were taken to ensure a higher follow-up rate?

- Quit attempt definition: We normally use the question, did you make a quit attempt lasting 24 hours or more as an indicator of a serious quit attempt. Has your question been validated? I also think future quit attempts is a bit confusing. These are smokers at baseline who made quit attempts during the study time period and reported them at 1 year follow-up. I might just say quit attempts at 1 year follow-up

- It seem like all of the smokers who were followed-up were still smoking at 1 year – were there any smokers who were abstinent?

- I appreciated that the authors presented the analysis using the Bayes factor and that does put the non-significant findings in perspective, however I do think the authors can clarify this section in the methods. For example, you could put the parameters for the BF for null hypothesis being true (BF close to 0), alternative hypothesis possible (BF close >1), and data insensitive (BF close to 1). This will allow for more clear interpretation of data in the results.

Results and conclusions

- I understand that the bayes factor for the interaction analysis showed that the data were insensitive to detect an association by SEP. I think it would help the reader to describe in more detail what the implications are of BF results that the data are insensitive to the association. From my understanding, it essentially means that a negative finding in your study does not rule out the possibility of an association in another study with a larger sample size and that this question merits consideration in larger studies. Perhaps including this in the first paragraph of discussion could be helpful.

- My main concern is around the follow-up sample being quite different from the overall sample at baseline. Therefore the conclusion of smoking for stress and quit attempt may only be relevant to a sub-sample that were followed up: older smoker, men, and those in higher SES groups. This does bring in some limitations to the conclusions – and not only were the results statistically non-significant but if they were significant then the might only apply to a sub-sample of smokers who were present at follow-up. I might make this more clear in the main findings of the study.

- Lastly, these were smokers who attempted to quit but were not successful so it does bring up the question about quit attempts being the mediator of successful quitting among those who smoke for stress. I.e., if we can demonstrate a connection that people who smoke for stress do attempt to quit and make multiple quit attempts, then that might be a marker of future successful quitting. Perhaps that point in the discussion could help frame these findings and provide a motivation for continuing to explore this association.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: David Thomas

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

25 Feb 2022

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

The manuscript now aligns with PLOS ONE’s style requirements.

2. Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified what type you obtained (for instance, written or verbal, and if verbal, how it was documented and witnessed). If your study included minors, state whether you obtained consent from parents or guardians. If the need for consent was waived by the ethics committee, please include this information.

Once you have amended this/these statement(s) in the Methods section of the manuscript, please add the same text to the “Ethics Statement” field of the submission form (via “Edit Submission”).

For additional information about PLOS ONE ethical requirements for human subjects research, please refer to http://journals.plos.org/plosone/s/submission-guidelines#loc-human-subjects-research.

We have now added the following to the “Ethics statement” on p.5 and the “Ethics approval and consent to participate” statement on p. 14:

“Respondents provided verbal informed consent prior to responding to the survey.”

3. Please include additional information regarding the survey or questionnaire used in the study and ensure that you have provided sufficient details that others could replicate the analyses. For instance, if you developed a questionnaire as part of this study and it is not under a copyright more restrictive than CC-BY, please include a copy, in both the original language and English, as Supporting Information. If the original language is written in non-Latin characters, for example Amharic, Chinese, or Korean, please use a file format that ensures these characters are visible.

We have included in the main body of the manuscript the exact wording of the survey items and response options used in the study.

4. Please state whether you validated the questionnaire prior to testing on study participants. Please provide details regarding the validation group within the methods section.

We have included details regarding the validation of the Motivation To Stop Scale on p.6-7. The remaining survey items are standard questions that have not all undergone formal validation but have been used in used in numerous papers that have been able to detect variations in quit attempt rates associated with mass media campaigns, and other population-level influences, as well as being the basis for the real-world evaluation of treatments, which in most cases replicated the results from RCTs.

https://pubmed.ncbi.nlm.nih.gov/31418449/

https://pubmed.ncbi.nlm.nih.gov/24322004/

https://academic.oup.com/ntr/article/22/9/1453/5524245?login=true

https://pubmed.ncbi.nlm.nih.gov/24372901/

5. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match. When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

We have now added a ‘Funding information’ section to the manuscript, which matches the ‘Financial disclosure’ section.

6. Thank you for stating the following in the Competing Interests section:

"OP, MT, and LK have no conflicts of interest to declare. SC has provided expert consultancy to providers of UK life insurance and the pharmaceutical industry on matters relating to smoking cessations aids. LS has received a research grant and honoraria for a talk and travel expenses from manufacturers of smoking cessation medications (Pfizer and Johnson & Johnson). JB has received unrestricted research funding from Pfizer to study smoking cessation." We note that you received funding from a commercial source: 'Pfizer and Johnson & Johnson.' Please provide an amended Competing Interests Statement that explicitly states this commercial funder, along with any other relevant declarations relating to employment, consultancy, patents, products in development, marketed products, etc. Within this Competing Interests Statement, please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared. Please include your amended Competing Interests Statement within your cover letter. We will change the online submission form on your behalf.

This has now been amended and added to the cover letter:

“OP, MT, SC and LK have no conflicts of interest to declare. LS has received a research grant and honoraria for a talk and travel expenses from manufacturers of smoking cessation medications (Pfizer and Johnson & Johnson). JB has received unrestricted research funding from Pfizer to study smoking cessation. This does not alter our adherence to PLOS ONE’s policies on sharing data and materials.”

7. Please note that in order to use the direct billing option the corresponding author must be affiliated with the chosen institute. Please either amend your manuscript to change the affiliation or corresponding author, or email us at plosone@plos.org with a request to remove this option.

I’ve asked the PLOS ONE team to merge multiple accounts (required to amend this information).

8. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

I’ve asked the PLOS ONE team to merge multiple accounts (required to amend this information).

9. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

This has now been amended.

Additional Editor Comments (if provided):

The authors report results of an interesting observational study examining the relations between self-reported stress-relief smoking motives and motivation to stop smoking and follow-up quit attempts. In particular, the authors are interested in whether SEP moderates these associations. These represent important scientific questions to inform movement toward achieving health equity in smoking treatment and intervention. The authors provide a detailed pre-registration on OSF in line with open science principles. Two expert reviewers provided their thoughtful critiques and recommendations below. I would encourage the authors to address their comments in detail, which I would consider a major revision.

We thank the Editor and the two reviewers for their helpful feedback on our manuscript.

The primary project seems well-designed, although additional methods details requested by both reviewers would enhance the manuscript and aid in interpretation of the data.

We have now added methodological details about the survey items, sampling strategy and incentivisation in the Methods section.

One significant concern is the very small sample size at follow up and in particular that this sample may not be representative of the larger initial sample (e.g., especially on the SEP moderator variable of interest). As reviewer 1 notes, the limitations should not simply be listed but rather discuss how they constrain the interpretation of the results. For the small follow up sample size and representativeness, this should especially be discussed in greater detail and provide more cautious interpretation of this follow up data given these concerns (see reviewer 2).

We provide a more elaborate discussion as to how the small sample size and lack of representativeness constrain the interpretation of the results on p.14.

Furthermore, I would encourage the authors to be careful with their language in the discussion to how they interpret non-significant effects per reviewer 1’s recommendations. Additionally, the description of the results in the discussion should mirror the findings in the results section – it is not appropriate to imply group differences when the data analyses do not reveal significant differences (e.g., discussion first paragraph, last sentence).

We have now modified the framing of the non-significant results throughout as per reviewer 1’s suggestion and have removed the sentence alluding to group differences in the Discussion on p.12.

Reviewer 1

This is a well-designed study with interesting results for their well-justified research question. However, the paper would be greatly improved by the authors paying more attention to the interpretation of their results in the Discussion.

We thank the reviewer for their constructive feedback on our manuscript, which we believe has strengthened it.

1. L33-42. Abstract. Revise the Results and Conclusion after addressing the suggestions below.

This has now been amended:

“Smoking to relieve stress is a commonly endorsed motive and is associated with high motivation to stop but not significantly with the odds of making a future quit attempt, although the magnitude and direction of the effect was similar for both outcomes. There was no clear evidence of moderation by SEP, although data were insensitive to distinguish the alternative from the null hypothesis.”

2. The introduction is clear and sets the seen well and justifies the research question well.

Thank you.

3. L188-190. Methods. It appears CPD has been treated as an interval variable rather than as a categorical variable, as for quit attempts in past year. This makes assumptions about the relationship between CPD and the outcomes which may not be justified, and a categorical variable may be more appropriate (and indeed they could consider using the combination measure of dependence Heaviness of Smoking index based on CPD and TTFC).

As suggested, we have conducted sensitivity analyses with CPD entered as a categorical variable, with associations remaining robust (p.10):

“In unplanned sensitivity analyses with CPD coded as a categorical variable, the associations were similarly unchanged (see S1 Table 2).”

4. L198. It appears just the baseline was weighted to the population, not the cohort that was followed up. Please clarify, and explain why the cohort was not weighted if it was not.

As is common within the STS, we did not observe important differences between the unweighted and weighted baseline samples (see Table 1). Therefore, the inferential analyses remain based on unweighted data, as planned. We also provide details on the unweighted and weighted follow-up sample in Table 1 (with differences between these being similarly small).

5. L211-214. The description of the alternative SEP variable based on housing tenure would be better moved to paragraph describing covariates (ending L190).

This has now been moved, as suggested.

6. L231-234. Results. This sentence only refers to p-values, not effect size or interval estimates. The authors should review how they report their results, especially results that are ‘not statistically significant’ in light of recommendations such as these and many more: Wasserstein RL, Lazar NA. The ASA statement on p‐values: context, process, and purpose. Am Stat 2016; 70: 129–133. Greenland S, McShane B. Scientists rise up against statistical significance. Nature 2019; 567: 305–307.

We agree with this (which is – we hope – at least partly reflected by us opting to complement the frequentist analyses with the calculation of Bayes Factors) and have reframed the reporting of the non-significant results throughout – e.g., in the Abstract:

“Smoking to relieve stress is a commonly endorsed motive and is associated with high motivation to stop but not significantly with the odds of making a future quit attempt, although the magnitude and direction of the effect was similar for both outcomes. There was no clear evidence of moderation by SEP, although data were insensitive to distinguish the alternative from the null hypothesis.”

7. L244-248. The OR of the interaction term is not informative to readers. The authors could consider reporting separately the ORs for high and low SEP subgroups, perhaps in a supplementary table, to demonstrate the magnitude and direction of the effect modification (even though it is not statistically significant, and the Bayesian analysis suggested it was insensitive to detect an interaction effect).

We now report the ORs for the higher and lower SEP groups separately in S1 Tables 3a and 3b (referred to on p.10).

8. L249-255. This paragraph would more logically sit before the paragraph above about interaction.

This has now been amended.

9. L251-255. Be more precise: start the sentence with ‘The adjusted association between the exposure and both outcome variables was unchanged…’ not ‘Results were materially …’. Drop the next two sentences as not germane, they are just about the adjusted association between an alternative covariate and the outcome.

This has now been amended.

10. Table 2. Consider just reporting an overarching p-value for the age and quit attempt categorical variables, rather than p-values for each category.

Thank you – as the relationships appear non-linear, we prefer to present the ORs and p-values for each category.

Should also include the n for each analysis, and a footnote explaining what is adjusted for in the adjusted ORs – all variables.

We now report the N for each analysis and include a footnote which indicates the covariates adjusted for.

Consider also reporting the unadjusted association of the co-variates with the outcomes.

We now report the unadjusted associations of the covariates with the outcomes in Table 1.

Please check the journal guidelines if the precision of the reported p-values is correct: typically I would report 0.08 (not 0.084), and only go to 3 decimal points if less than 0.01.

PLOS ONE’s guidelines only state that exact p-values (if greater than 0.001) should be reported but do not mention the number of decimal points.

11. Supplementary table – as above.

This has now been amended, as suggested.

12. Discussion. Principal findings. Second and third sentence. Reconsider how you report non-significant findings. This does not mean that there is no association, just that you were unable to detect one. The point estimate of the association with making a quit attempt is of similar magnitude and direction as the association with motivation – however this is based on a smaller sample with less power. So don’t assume no impact, indeed consider discussing WHY there may also be an impact on quit attempts. The last sentence of this paragraph is inconsistent with the results – there is small non-significant difference between the groups: they are similar.

We have now changed the wording pertaining to the non-significant results on p.12. We agree that the last sentence is inconsistent with the results and have removed it.

“Although non-significant, both the direction and magnitude of the association of smoking to relieve stress with future quit attempts were similar to those for the association with high motivation to stop. We did not observe significant moderation by SEP; however, the Bayes Factors indicated that data were insensitive to detect an association (as opposed to lending support to the null hypothesis of no association). For this to be known, other studies with larger samples need to be conducted to detect any potential moderation effects by SEP.”

13. Discussion. Principal findings. Second paragraph. There is only scant discussion of possible reasons for the authors’ principal findings. The smoker’s belief that they are addicted is true, but the belief that they are motivated to stay smoking because they believe smoking helps them cope with stress is based on a misperception. The authors get close to explaining the link in the fourth sentence, but this needs to be stepped out more clearly. This should come before the previous sentence as that is about a methodological reason why this association may be spurious due to reverse causation. Might there also be a common factor that causes both the explanatory and outcome factor?

Thank you – we have now moved the methodological consideration about reverse causation to the end of the paragraph and included further discussion of possible reasons for the principal results.

“First, people who smoke to relieve stress may evaluate their smoking experience as biological or instrumental (as opposed to pleasurable), which may be perceived as unpleasant or annoying, hence leading to greater motivation to stop. Second, those who smoke to relieve stress may find smoking enjoyable but at the same time, find it a source of dissonance to use smoking as a ‘crutch’ to cope with stress, hence leading to greater motivation to stop. Third, it is also possible that there is a common factor (e.g., depression) that gives rise both to the belief that smoking helps to relieve stress and high motivation to stop. Previous research has found that smokers with more severe (compared with less severe) levels of depression report lower quitting self-efficacy (which may be influenced by positive outcome expectations, such as the belief that smoking helps to cope with stress) but higher motivation to stop [33].”

The authors need to also provide some explanation for other principal findings, e.g. no difference in the belief that they are motivated to stay smoking because they believe smoking helps them cope with stress in the two SEP groups, before jumping to the need for more research, strengths and limitations, and implications for practice. These explanations need to better contextualise their new results with previous research.

Thank you – we now discuss this on p.13-14:

“The finding that smokers from both higher and lower SEP groups endorse the belief that smoking helps to relieve stress may indicate that smokers are not only more/less likely to endorse this belief depending on their level of actual stressors (which may, on average, be higher in smokers from lower SEP groups). Instead, this may be related to people’s perceived ability to cope with stressors. Smokers across the social spectrum may perceive themselves to be better/worse equipped to cope with stress (irrespective of whether they actually are or not), thus influencing their beliefs about the role of smoking (e.g. positive outcome expectancies).”

14. Discussion. Strengths and limitations. Rather than just list these, the authors need to consider how they are likely to have altered their results and interpretation and implications of the results.

Thank you – we now elaborate on this in the Discussion on p.14:

“Second, the follow-up sample differed from the baseline sample on important characteristics, including SEP. This means that, had our findings pertaining to the follow-up sample been statistically significant, they may not be generalisable to the general population of smokers. In addition, if those who were more likely to be followed up (i.e. older smokers, men, those from higher SEP groups) were also more or less likely to endorse the belief that they smoke to relieve stress, this may have distorted any association with the outcome. Third, although the Smoking Toolkit Study is a nationally representative survey on key sociodemographic and smoking characteristics, it may not be representative with regards to psychological characteristics such as motivation to stop or the motive that smoking helps to relieve stress and it also may not reach smokers who are the most disadvantaged. Similar to the previous point about the follow-up sample, this may mean our findings are not generalisable to the general population of smokers.”

15. Discussion. Implications. First paragraph, second sentence. But if the perception is associated with increased motivation (and possibly increased likelihood of making a quit attempt in the future), this may not be a problem that needs to be corrected! Please discuss.

We now discuss this on p.15:

“Although the finding that smoking to relieve stress is associated with high motivation to stop (and possibly the likelihood of making a quit attempt in the future) may be perceived as a positive finding that does not warrant intervention, it may be more ethically defensible to highlight to smokers the large body of work indicating that smoking cessation leads to positive mental health outcomes and use this as a motive for stopping, thus correcting common misperceptions. In addition, if smoking is normatively perceived as a stress reducer, then it may encourage people to take up smoking. This merits further investigation.”

In contrast, in the next sentence, this same misperception being held by health professionals definitely is a problem (which the authors have not measured or examined) and needs to be countered as it leads to less support. There is now increasing research evidence about how best counter such misperceptions without inadvertently reinforcing them, eg about vaccines and climate change, that could be referred to here.

We have added the following on p.15:

“At the same time (although not measured in the present study), research shows that many mental health professionals believe that smoking helps to alleviate stress and therefore refrain from offering appropriate smoking cessation support to their clients [38]. This suggests that the tackling of common misperceptions also needs to occur among practitioners. Public health campaigns have often focused on the well-documented physical health aspects of smoking, with less attention paid to the mental health benefits of quitting, including reduced stress, anxiety and depression [6] but also increased levels of happiness [39,40]. With increasing evidence for how to best counter misperceptions without inadvertently reinforcing them in the areas of vaccine hesitancy and climate change (e.g. normative feedback, framing messages in terms of gain) [41-43], interventions targeting healthcare professionals may benefit from drawing on this growing literature.”

16. Discussion. Implications. Second paragraph. The findings suggest that addressing misperceptions will be equity neutral, as similar association in high and low SEP. The paper does NOT address the association between stress and quitting, there are no implications from the results about addressing stress itself.

We agree that this paragraph goes beyond the current findings and have therefore removed it.

17. Discussion. Implications. Last paragraph. The first sentence is well substantiated, but the second sentence is not substantiated. Indeed, from other research, motivation seems mainly associated with initiating but not sustaining quit attempts.

Thank you – this has now been amended to reflect the evidence:

“It is plausible that people from lower SEP groups endorse a confluence of motivational, cognitive and social factors for continued smoking (e.g. smoking to relieve stress and boredom, lower quitting self-efficacy, social aspects of smoking), which may interact to make it more difficult to quit [45]. Future research should examine the potential clustering of such factors and whether certain patterns are more pronounced in lower (compared with higher) SEP groups, as this is needed to help reduce smoking-related health inequalities.”

18. Conclusion. The second sentence is correct and is a more correct description of the last clause of the first sentence.

We have now amended this, as follows:

“Smoking to relieve stress is a commonly endorsed motive and is associated with high motivation to stop but not significantly with the odds of making a future quit attempt, although the magnitude and direction of the effect was similar for both outcomes. There was no clear evidence of moderation by SEP, although data were insensitive to distinguish the alternative from the null hypothesis.”

Note that I am insufficiently familiar with Bayesian statistics to adequately review those methods.

Thank you.

Reviewer 2

The authors have presented an interesting manuscript that explored the relationship between using smoking to relieve stress and motivation to quit and future quit attempts in a UK sample of people 16 years and older. I think the authors did a good job laying out the premise for the analysis and I do think that there is merit in understanding how stress and coping with stress impacts smoking behaviors. The relationship is particularly salient among low SES populations as often we are inclined to advise people to quit but we don’t offer strategies to help them with the many stressors that lead them to smoke in the first place. I did have some questions about the methods and conclusions, which I will highlight below.

We thank the reviewer for their helpful feedback on our manuscript, which we believe has strengthened it.

Methods

- Providing additional details on the Smoking Toolkit Study or a reference would be helpful. For example, I had the following questions:

o The sample was hybrid of probability and quota sample. Could you share how much was a probability based sample?

We have now provided more information about the sampling strategy in the Methods on p.5, which clarifies why we cannot estimate how much was a probability-based sample:

“Interviews are held with one household member in selected geographic output areas until quotas are fulfilled. The quotas are based on factors influencing the probability of being at home (i.e. working status, age and gender). This hybrid form of random probability and quota sampling is considered superior to conventional quota sampling. Here, the choice of households to approach is limited by the random allocation of small output areas and rather than being sent to specific households in advance, interviewers can choose which households within these small geographic areas are most likely to fulfil their quotas. Therefore, unlike random probability sampling, it is not appropriate to record the response rate in the STS.”

o Were respondents to the survey incentivized?

We have now clarified this on p.5:

“Follow-up respondents were not remunerated.”

o How many waves are there for this survey?

We have now clarified the number of waves on p.5:

“Data included in the present study were collected across four waves (139-142) from respondents surveyed between April and July 2018 (when smoking to relieve stress was included in the STS).”

o How were attempts made to reconnect with survey participants at 1 year follow-up? Why were there only 153 people at follow-up from such a large sample at baseline. Was the survey designed to be longitudinal? What steps were taken to ensure a higher follow-up rate?

We have now clarified this on p.5:

“Follow-up respondents are not remunerated. Up to eight attempts are made to follow up each consenting respondent, with calls made at different times of day across weekdays and weekends.”

We have also clarified in the Discussion on p.13-14 that the small sample size for the second research question was at least partly due to limited funding:

“First, the sample size for the prospective association of smoking to relieve stress and future quit attempts was small. However, this was pre-determined by the period for which the funding was available (i.e., the question about smoking motives – including smoking to relieve stress – was included only in a few STS waves)…”

- Quit attempt definition: We normally use the question, did you make a quit attempt lasting 24 hours or more as an indicator of a serious quit attempt. Has your question been validated?

Although this survey item has not been formally validated, the question was cognitively tested when establishing the survey in 2006, and has been used in numerous papers that have been able to detect variations in quit attempt rates associated with mass media campaigns, and other population-level influences, as well as being the basis for the real-world evaluation of treatments, which in most cases replicated the results from RCTs.

https://pubmed.ncbi.nlm.nih.gov/31418449/

https://pubmed.ncbi.nlm.nih.gov/24322004/

https://academic.oup.com/ntr/article/22/9/1453/5524245?login=true

https://pubmed.ncbi.nlm.nih.gov/24372901/

I also think future quit attempts is a bit confusing. These are smokers at baseline who made quit attempts during the study time period and reported them at 1 year follow-up. I might just say quit attempts at 1 year follow-up

This has been amended throughout.

- It seem like all of the smokers who were followed-up were still smoking at 1 year – were there any smokers who were abstinent?

We did not examine the smoking status of those followed up at 12 months as this was not one of our research questions.

- I appreciated that the authors presented the analysis using the Bayes factor and that does put the non-significant findings in perspective, however I do think the authors can clarify this section in the methods. For example, you could put the parameters for the BF for null hypothesis being true (BF close to 0), alternative hypothesis possible (BF close >1), and data insensitive (BF close to 1). This will allow for more clear interpretation of data in the results.

We have now clarified the interpretation of the Bayes Factors on p.8:

“Bayes Factors (BFs) can be interpreted as follows: BFs >3 lend support to the alternative hypothesis (over the null), BFs <1/3 lend support to the null, and BFs in-between reflect data being insensitive to distinguish the two.”

Results and conclusions

- I understand that the bayes factor for the interaction analysis showed that the data were insensitive to detect an association by SEP. I think it would help the reader to describe in more detail what the implications are of BF results that the data are insensitive to the association. From my understanding, it essentially means that a negative finding in your study does not rule out the possibility of an association in another study with a larger sample size and that this question merits consideration in larger studies. Perhaps including this in the first paragraph of discussion could be helpful.

Thank you for this helpful suggestion – we have now added this on p.13:

“We did not observe significant moderation by SEP; however, the Bayes Factors indicated that data were insensitive to detect an association (as opposed to lending support to the null hypothesis of no association). This means that it is plausible that another study with a larger sample may pick up moderation effects by SEP (if they exist).”

- My main concern is around the follow-up sample being quite different from the overall sample at baseline. Therefore the conclusion of smoking for stress and quit attempt may only be relevant to a sub-sample that were followed up: older smoker, men, and those in higher SES groups. This does bring in some limitations to the conclusions – and not only were the results statistically non-significant but if they were significant then the might only apply to a sub-sample of smokers who were present at follow-up. I might make this more clear in the main findings of the study.

Thank you – we now elaborate on this in the Discussion on p.14:

“Second, the follow-up sample differed from the baseline sample on important characteristics, including SEP. This means that, had our findings pertaining to the follow-up sample been statistically significant, they may not be generalisable to the general population of smokers. In addition, if those who were more likely to be followed up (i.e. older smokers, men, those from higher SEP groups) were also more or less likely to endorse the belief that they smoke to relieve stress, this may have distorted any association with the outcome.”

- Lastly, these were smokers who attempted to quit but were not successful so it does bring up the question about quit attempts being the mediator of successful quitting among those who smoke for stress. I.e., if we can demonstrate a connection that people who smoke for stress do attempt to quit and make multiple quit attempts, then that might be a marker of future successful quitting. Perhaps that point in the discussion could help frame these findings and provide a motivation for continuing to explore this association.

Thank you – we did not examine respondents’ smoking status at the 12-month follow-up (only if they had attempted to stop). However, we believe this is a useful point to highlight and have added the following in the Discussion on p.14:

“As many quit attempts are, on average, needed before quitting smoking successfully [36], a positive relationship between smoking to relieve stress and making a quit attempt in the future would likely have a positive impact on public health.”

Submitted filename: Responses to reviewer comments 25.02.22.docx

Click here for additional data file.

2 May 2022

Associations between smoking to relieve stress, motivation to stop and quit attempts across the social spectrum: A population survey in England

PONE-D-21-18453R1

Dear Dr. Perski,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Jesse T. Kaye, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Thank you for resubmitting this revised manuscript for publication at PLOS ONE. The authors responded very thoroughly to the reviewers initial round of comments and the manuscript is improved as a result. Methods are explained in greater detail and the discussion of study findings is more comprehensive and also appropriately constrained by the data analysis and study limitations. This will be an important contribution to the literature. One of the original reviewers was able to review the revised manuscript and offered the suggestion to include the omnibus p-value for the categorical variables with multiple groups (age and number of quit attempts). The authors may add this to the table during copy editing if they would like (in addition, not instead of the individual p values). I agree it would enhance the table, but not absolutely required. Thank you for conducting this research to better understand the relations between stress coping motives and smoking cessation motivation.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: #10. Table 2. I agree relationships non-linear and a categorical variable based on series of dummy dummy variables is appropriate. I do not suggest a continuous variable. However most statistical programs (eg Stata) enable a post-estimation command which provides a p-value for the addition of the overall categorical variable. Not just the individual p-values comparing categories to the reference. I suggest this be included not the individual p-values.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

9 May 2022

PONE-D-21-18453R1

Associations between smoking to relieve stress, motivation to stop and quit attempts across the social spectrum: A population survey in England

Dear Dr. Perski:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Jesse T. Kaye

Academic Editor

PLOS ONE