Abstract
Background
Social norms have received little attention in relation to electronic cigarettes (EC). The current study examine social norms for EC use and smoking tobacco, and their associations with (i) initiation of EC use, (ii) intention to quit smoking and (iii) attempts to quit smoking.
Methods
Cross-sectional and longitudinal data analysis from Waves 1 and 2 of the ITC 6 European Country Survey and corresponding waves from England (the ITC Four Country Smoking and Vaping Survey). Current smokers at baseline, who heard of ECs and provided data at both waves were included (n = 3702). Complex samples logistic regression examined associations between the outcomes and descriptive (seeing EC use in public, close friends using ECs/smoking) and injunctive (public approves of ECs/smoking) norms, adjusting for country, demographics, EC use and heaviness of smoking.
Results
In longitudinal analyses, seeing EC use in public at least some days was the only social norm that predicted initiation of EC use between waves (OR = 1.66, 95%CI = 1.08–2.56). In the cross-sectional analysis, having an intention to quit was associated with seeing EC use in public (OR = 1.37, 95%CI = 1.04–1.81) and reporting fewer than three close friends smoke (OR = 0.59, 95%CI = 0.44–0.80). There was no association between any social norm and making a quit attempt between waves.
Conclusions
Initiation of EC use is predicted by seeing EC use in public, which was also associated with greater intention to quit smoking. Friends’ smoking was associated with lower intention to quit. These findings may allay concerns that increased visibility of ECs is renormalizing smoking amongst current smokers.
Introduction
The use of electronic cigarettes (ECs) has increased across Europe in recent years.1 Current prevalence estimates of those aged 15 and over report that 9% have tried using ECs once or twice and 2% currently use ECs in the European Union (EU). Within the UK, 10% have tried using ECs and 5% are current users.2 As the number of those using ECs has increased, so have debates regarding the potential harms and benefits of ECs; some reports3–5 have concluded that ECs carry significantly lower risk than smoking tobacco (referred to as ‘smoking’ from this point forward) and may aid cessation, while others have raised concerns about the unknown short- and long-term health effects.6 Given that ECs can be considered as potentially harm-reducing for current smokers; this provides an argument for identifying key modifiable predictors for EC use.
Social norms are frequently investigated as predictors of health behaviours. These variables are also key components of models of behaviour, such as the Theory of Planned Behaviour (TPB).7 This model includes 'injunctive' social norms, which represent an evaluation of whether an individual feels that significant others or society as a whole thinks he/she should engage in the behaviour.7 As per the TPB, the more an individual feels that others approve of EC use, the more likely they are to intend to use ECs. In turn, this increases the likelihood of EC use. ‘Descriptive’ social norms have also been examined as distinct predictors of behaviour.8 Descriptive social norms are beliefs about how others behave, e.g. the numbers of close friends, family members or society as a whole who use ECs. In addition to influencing behaviour through motivation, the social environment within which a behaviour takes place can also facilitate the uptake or maintenance of behaviour, e.g. by providing cues or opportunities for smoking to take place.9
Cognitive measures of injunctive and descriptive social norms of smoking have been found to predict smoking onset, intentions to quit and attempts to quit smoking.10–14 Furthermore, having friends who smoke has frequently been found to predict onset of smoking in adolescents15 and a reduction in the number of close friends who smoke has been found to be associated with greater likelihood of quit attempts and of quit success.16 Unsurprisingly, given the relatively recent emergence of EC use, research into associations between social norms and EC use is rarer, although results thus far suggest a similar pattern. Two longitudinal studies amongst youth found having friends who use ECs predicted subsequent initiation of EC use.17,18 One of these also found that perceiving that one’s best friend would react positively to EC use were associated with increased odds of subsequent EC initiation.18 Cross-sectional data have also found that the number of friends and family using ECs differs among smokers, EC users and those who use both products.19
There are also data on the associations between EC use and social norms for smoking, and conversely, between smoking and social norms for EC use. In youth populations, trial of ECs has been predicted by having friends who smoke.20 The impact of EC social norms on smoking behaviour is of particular interest as concern has been raised that EC use has the potential to ‘renormalize’ smoking, e.g. if EC use becomes more accepted and visible in society, then by comparison smoking may also seem more acceptable21,22 which could deter motivation and attempts to quit smoking. Whether this is occurring has not yet been established. A study conducted amongst youth found that having friends who use ECs predicted initiation of smoking.17 However, a study conducted amongst adults23 found no evidence that those regularly exposed to EC use were less motivated to quit smoking or less likely to make a quit attempt than those who were not regularly exposed.
Research questions
The current study analyses longitudinal data from Waves 1 and 2 of the International Tobacco Control Project 6 European Country Survey (ITC 6E, Germany, Greece, Hungary, Poland, Romania and Spain)24 and Waves 1 and 2 England data of the ITC Four Country Smoking and Vaping (ITC 4CV) Survey.25 Three research questions were addressed. First (i), the study examines whether descriptive and injunctive social norms for EC use and smoking, measured at Wave 1, predict initiation of EC use at Wave 2. Second (ii), we examine cross-sectional associations between descriptive and injunctive social norms for EC use and smoking and the intention to quit smoking, measured at Wave 1. Finally (iii), we will examine whether social norms for EC use and smoking, measured at Wave 1, predict attempts to quit smoking at Wave 2.
Methods
Participants and design
The current study is part of the Horizon 2020-funded EUREST-PLUS (European Regulatory Science on Tobacco: Policy implementation to reduce lung diseases) project (EUREST-PLUS-HCO-06-2015). The follow-up period for both studies was between 15 months and 2 years. Follow-up Wave 1 of the English data from ITC 4CV used in the study was collected from July to November 2016 and Wave 2 data were collected between February and July 2018. The European data from ITC 6E were collected from June to September 2016 and February to May 2018, respectively. The sample size at Wave 1 of the ITC 6E survey was 6011, 3195 (69.6%) of whom completed survey at both waves. The sample size at Wave 1 of the ITC 4CV survey was 3536, 1394 (39.4%) of whom completed survey at both waves. Of the 4589 respondents from both surveys who provided data at each wave, in this analysis, we included data from 3702 (80.7% of those with available data at both waves) who reported that they had heard of ECs at Wave 1, as this item was a pre-requisite for being asked to respond to EC social norm items. Further study and participant details are provided in Thompson et al.25.
Ethical clearance
The survey protocols and all materials, including the survey questionnaires, were cleared for ethics by the Office of Research Ethics, University of Waterloo, Canada and by local ethics boards in each study country. All respondents provided consent to participate.
Outcome measures
All variables were binary, or recoded as binary, except where indicated.
Initiation of EC use between Waves 1 and 2
At Waves 1 and 2, respondents were asked ‘Have you ever used an e-cigarette or vaping device, even one time?’ with responses: yes/no. For research question (i) only those responding ‘no’ at Wave 1 were included. Those responding ‘yes’ at Wave 2 were coded as having initiated EC use.
Intention to quit smoking
One item assessed intention to quit smoking at Wave 1: ‘Are you planning to quit smoking…’ with responses classified into: planning to quit in the next 6 months (within the next month; within the next 6 months) vs. all other responses (sometime in the future, beyond 6 months; not planning to quit, do not know).
Quit attempts
Attempts to quit smoking were assessed at Wave 2 using the item: ‘Have you made any attempts to stop smoking since we last spoke to you?’ with responses yes/no.
Predictor variables
Wording for certain items differed between the ITC 6E and ITC 4CV surveys. All social norms were recoded as dichotomous variables.
Social norms for EC use
Two items assessed descriptive social norms for EC use. The first, ‘friend EC use’, was assessed using different wording in the ITC 6E and ITC 4CV surveys. In the ITC 6E survey, Friend EC use was assessed with the item: ‘Of the five closest friends or acquaintances that you spend time with on a regular basis, how many of them use e-cigarettes/vaping devices?’ with responses 0–5. In the ITC 4CV survey, Friend EC use was assessed using two items. First, respondents were asked ‘How many friends or acquaintances do you spend time with on a regular basis?’ with responses 0–5, More than 5. This was followed by ‘Of (these 1-5/the 5 closest) friends or acquaintances that you spend time with on a regular basis, how many of them use e-cigarettes/vaping devices?’ with responses 0–5. In both surveys, response options were recoded as none or ≥1. Those responding ‘don’t know’ were counted as missing data and excluded from the analysis. 'Seeing EC use in public' was also assessed using different wording in the ITC 6E and ITC 4CV surveys. In the ITC 6E survey, seeing EC use in public was assessed using the item: ‘In the last 30 days, how often have you seen anyone using an e-cigarette or vaping device in public?’. In the ITC 4CV survey, seeing EC use in public was assessed using the item ‘In the last 30 days, how often, if at all, have you seen anyone vaping (using e-cigarettes) in public?’. In both cases, responses were recoded as at least some days (every day, most days, some days) vs. all other responses (rarely, not at all or do not know).
One item assessed injunctive social norms. In the ITC 6E survey, this was assessed using the item: ‘What do you think the general public’s attitude is towards using e-cigarettes or vaping devices?’. In the ITC 4CV survey, this was assessed using ‘What do you think the general public’s attitude is towards vaping/using e-cigarettes' (ITC 4CV). Responses were recoded as approves (strongly approves, approves) vs. does not approve (neither approves nor disapproves; somewhat disapproves; strongly disapproves, do not know).
Social norms for smoking
One item assessed descriptive social norms for smoking assessed using different wording in the ITC 6E and ITC 4CV surveys. In the ITC 6E survey, ‘Friend smoking’ was assessed a single item: ‘Of the five closest friends or acquaintances that you spend time with on a regular basis… How many of them smoke ordinary cigarettes?’, with responses 0–5. In the ITC 4CV survey, respondents were asked two questions ‘How many friends or acquaintances do you spend time with on a regular basis? 0–5, More than 5’, followed by ‘Of (these 1-5/the 5 closest) friends or acquaintances that you spend time with on a regular basis, how many of them smoke ordinary cigarettes? With responses 0–5’. As the prevalence of smoking cigarettes is higher than using ECs in the countries where data were collected, these were coded differently to the corresponding EC item as ≤2 or ≥3 friends who smoke. As above, those responding ‘don’t know’ were counted as missing data and excluded from the analysis. One item assessed injunctive social norms for smoking, assessed using the same item in both surveys, ‘What do you think the general public’s attitude is towards smoking cigarettes?’, recoded as approve (strongly approves, somewhat approves) or does not approve (somewhat disapproves, strongly disapproves, neither approves nor disapproves, do not know).
Covariates
Covariates included in all multivariable analyses were age, sex, income, education and country. Analyses examining the intention to quit smoking and attempts to quit smoking also controlled for the heaviness of smoking index (continuous variable with a range of 0–6)26 and current use of ECs (all respondents reporting current use of ECs, either daily, weekly or less than weekly/occasional were coded as current users and compared with all other responses). Analyses examining attempts to quit smoking further controlled for the intention to quit smoking. All covariates were assessed at Wave 1.
Analysis
Logistic regression analyses were conducted to examine: (i) prospective associations between descriptive and injunctive EC social norms at Wave 1 and the initiation of EC use (reporting having ever used ECs) at Wave 2; (ii) cross-sectional associations between descriptive and injunctive EC social norms at Wave 1 and the intention to quit smoking at Wave 1; and (iii) prospective associations between descriptive and injunctive EC social norms at Wave 1 and attempts to quit smoking at Wave 2. Analyses for (i) were restricted to those who reported never having used ECs at Wave 1 and who provided a response to the item on ever use of ECs at Wave 2. For each analysis, both unadjusted and adjusted (see covariates section, above) complex samples logistic regression analyses were conducted. We calculated and reported odds ratios and corresponding 95%CI for each predictor variable. Odds ratios for individual variables were adjusted for all other variables in each model. Longitudinal (for analyses (i) and (iii)) and cross-sectional for (ii) sampling weights and strata will be applied in each model. Complete case analysis was used. Cases where respondents refused to answer an item were always counted as missing data. All analyses were conducted using IBM SPSS 25.
Results
Participant characteristics at baseline for the two analytic samples are shown in table 1. England provided the largest proportion of respondents for the cross-sectional and longitudinal datasets (29.0% and 37.0%, respectively). There was a majority of male participants in both datasets (55.0% and 56.2%, respectively) and most participants were in the 25–54 age range (63.6% and 63.2%, respectively). The vast majority of respondents in both datasets were daily smokers (93.4 and 92.7, respectively).
Table 1.
Baseline characteristics in 2016 of current smokers from seven European countries reporting having heard about EC and who contributed data to samples for research question (i) (n = 1758) and (ii) and (iii) (n = 3702)
| Subsample of current smokers who heard about ECs but have not used ECs at baseline (n = 1758) (Research question i) |
Total sample (n = 3702) (Research questions ii and iii) |
|||
|---|---|---|---|---|
| Country | n a | %b | n a | %b |
| Germany | 262 | 14.0 | 452 | 12.1 |
| Greece | 161 | 9.9 | 280 | 8.0 |
| Hungary | 149 | 7.9 | 251 | 6.7 |
| Poland | 153 | 8.9 | 351 | 9.3 |
| Romania | 170 | 9.5 | 364 | 10.4 |
| Spain | 393 | 20.8 | 614 | 16.6 |
| England | 470 | 29.0 | 1390 | 37.0 |
| Gender | ||||
| Male | 883 | 55.0 | 1845 | 56.2 |
| Female | 875 | 45.0 | 1857 | 43.8 |
| Age | ||||
| 18–24 | 104 | 6.5 | 311 | 12.6 |
| 25–39 | 397 | 25.1 | 894 | 31.8 |
| 40–54 | 664 | 38.5 | 1291 | 31.4 |
| 55 and over | 594 | 29.9 | 32.6 | 24.2 |
| Monthly household income | ||||
| Low | 417 | 22.3 | 861 | 21.2 |
| Moderate | 620 | 33.9 | 1284 | 33.3 |
| High | 355 | 21.8 | 894 | 25.6 |
| Not stated | 366 | 22.0 | 663 | 19.9 |
| Highest level of education | ||||
| Low | 690 | 35.2 | 1342 | 32.3 |
| Moderate | 807 | 52.8 | 1672 | 55.7 |
| High | 248 | 12.0 | 661 | 12.0 |
| Smoking status | ||||
| Daily | 1655 | 93.4 | 3447 | 92.7 |
| Non-daily | 103 | 6.6 | 183 | 7.3 |
| EC use status | ||||
| Daily | 0 | 0 | 151 | 6.3 |
| Non-daily | 0 | 0 | 553 | 14.0 |
| No current use | 1758 | 100 | 2998 | 79.7 |
| Heaviness of smoking index [mean (SD)] | 2.35 (0.05) | 2.49 (0.04) | ||
| Intention to quit smoking at W1 | ||||
| Intends to quit | 267 | 15.1 | 737 | 19.4 |
| Does not intend to quit | 1490 | 84.9 | 2693 | 80.6 |
| Made an attempt to quit smoking (Wave 2) | ||||
| Yes | 300 | 19.7 | 1045 | 71.8 |
| No | 1457 | 80.3 | 2656 | 28.2 |
| Number of five closest friends who use ECs | ||||
| At least one | 216 | 13.9 | 700 | 20.3 |
| None | 1426 | 86.1 | 2704 | 79.7 |
| Frequency of seeing EC use in public | ||||
| At least some days | 587 | 35.1 | 1676 | 45.6 |
| All other responses | 1168 | 64.9 | 2022 | 54.4 |
| General public’s attitude towards ECs | ||||
| Approves | 409 | 24.5 | 989 | 25.9 |
| Does not approve | 1348 | 75.5 | 2712 | 74.1 |
| Number of five closest friends who smoke | ||||
| Three or more | 985 | 59.6 | 1896 | 56.6 |
| Two or fewer | 661 | 40.4 | 1527 | 43.4 |
| General public’s attitude towards smoking | ||||
| Approves | 206 | 11.8 | 416 | 11.8 |
| Does not approve | 1552 | 88.2 | 3286 | 88.2 |
aUnweighted.
bWeighted. Details of sample for research question (i): Longitudinal sample, restricted to those who reported never having used ECs at Wave 1 and who provided a response to the item on ever use of ECs at Wave 2. Details of sample for research questions (ii) and (iii): All respondents who completed surveys at both waves and reported that they had heard of ECs at Wave 1.
Prospective associations between EC social norms and the initiation of EC use
A total of 1758/3702 (47.5%) respondents reported that they had never used ECs at Wave 1 and provided data on ever use of ECs at follow-up. Of those reporting never having used ECs at Wave 1, 17.2% (n = 302) reported that they had initiated EC use (reported ever use) at Wave 2. Results of bivariate and multivariable analyses predicting initiation of EC use are shown in table 2. In multivariable analyses, only one social norms variable remained predictive; reporting seeing EC use at least some days increased the odds of having tried ECs at Wave 2 (OR = 1.66, 95%CI = 1.08–2.56, P = 0.020). Country and age also significantly predicted the initiation of EC use between waves (P < 0.001 and P = 0.001, respectively).
Table 2.
Prospective associations between descriptive and injunctive social norms at Wave 1 and the initiation of EC use at Wave 2 (n = 1758)
| Initiated EC use between Waves 1 and 2 |
Associations between predictor variables and initiating EC use between waves (n = 1758) |
|||
|---|---|---|---|---|
| Bivariate | Multivariable | |||
| n/N | % | OR (95%CI) | OR (95%CI) | |
| Number of five closest friends who use ECs | ||||
| At least one | 53/216 | 26.5 | 1.63 (1.01–2.62) | 1.38 (0.81–2.34) |
| None | 223/1426 | 18.1 | 1.00 | 1.00 |
| Frequency of seeing EC use in public | ||||
| At least some days | 159/587 | 31.0 | 2.87 (2.05–4.03) | 1.66 (1.08–2.56) |
| All other responses | 142/1168 | 13.5 | 1.00 | 1.00 |
| General public’s attitude towards ECs | ||||
| Approves | 85/409 | 24.4 | 1.45 (0.99–2.10) | 1.24 (0.80–1.92) |
| Does not approve | 217/1348 | 18.2 | 1.00 | 1.00 |
| Number of five closest friends who smoke | ||||
| Three or more | 140/985 | 16.3 | 0.61 (0.43–0.86) | 1.08 (0.69–1.68) |
| Two or fewer | 139/661 | 24.2 | 1.00 | 1.00 |
| General public’s attitude towards smoking | ||||
| Approves | 27/206 | 19.1 | 0.95 (0.52–1.75) | 1.03 (0.53–1.99) |
| Does not approve | 275/1552 | 19.8 | 1.00 | 1.00 |
| Country | ||||
| Germany | 45/262 | 20.7 | 0.50 (0.29–0.85) | 0.66 (0.34–1.25) |
| Greece | 35/161 | 18.6 | 0.44 (0.27–0.71) | 0.43 (0.23–0.84) |
| Hungary | 12/149 | 9.8 | 0.21 (0.10–0.44) | 0.28 (0.12–0.67) |
| Poland | 10/153 | 8.8 | 0.19 (0.08–0.42) | 0.19 (0.08–0.46) |
| Romania | 23/170 | 17.6 | 0.41 (0.22–0.76) | 0.47 (0.22–0.96) |
| Spain | 38/393 | 9.1 | 0.19 (0.12–0.31) | 0.25 (0.13–0.47) |
| England | 139/470 | 34.3 | 1.00 | 1.00 |
| Gender | ||||
| Female | 140/735 | 17.0 | 0.74 (0.53–1.02) | 0.86 (0.61–1.21) |
| Male | 162/721 | 21.7 | 1.00 | 1.00 |
| Age | ||||
| 25–39 | 87/397 | 23.7 | 0.50 (0.27–0.92) | 0.50 (0.26–0.97) |
| 40–54 | 113/664 | 16.3 | 0.31 (0.17–0.57) | 0.37 (0.20–0.69) |
| 55 and over | 74/593 | 13.3 | 0.25 (0.13–0.46) | 0.27 (0.14–0.52) |
| 18–24 | 28/104 | 38.4 | 1.00 | 1.00 |
| Annual income | ||||
| Low | 68/417 | 16.0 | 0.47 (0.30–0.74) | 0.93 (0.54–1.60) |
| Moderate | 116/620 | 21.0 | 0.66 (0.43–1.00) | 1.02 (0.63–1.66) |
| Not stated | 36/366 | 12.5 | 0.35 (0.19–0.65) | 0.80 (0.40–1.58) |
| High | 82/355 | 28.9 | 1.00 | 1.00 |
| Highest level of education | ||||
| Low | 116/690 | 18.40 | 0.84 (0.51–1.39) | 1.10 (0.61–1.97) |
| Moderate | 130/807 | 19.4 | 0.99 (0.55–1.46) | 1.00 (0.59–1.74) |
| High | 53/248 | 21.1 | 1.00 | 1.00 |
Numbers (n/N) are unweighted, percentages are based on weighted data; analysis only include those reporting never having used ECs at Wave 1 and who provided a response to the item on ever use of ECs at Wave 2. Significant associations are in bold.
Cross-sectional associations between EC social norms and the intention to quit smoking
The second set of analyses (table 3) examined the cross-sectional associations between EC and smoking social norms and intending to quit smoking within the next 6 months among all respondents. In multivariable analyses, only two social norm variables remained significant; reporting seeing EC use in public at least some days increased the odds of intending to quit (OR = 1.37, 95%CI = 1.04–1.81, P = 0.024), whereas reporting three or more of their close friends smoke decreased the odds (OR = 0.59, 95%CI = 0.44–0.80, P < 0.001). Intention to quit smoking was also significantly associated with country (P < 0.001), gender (P = 0.023) and age (P = 0.017).
Table 3.
Cross-sectional associations between descriptive and injunctive EC social norms at Wave 1 and the intention to quit smoking at Wave 1 (n = 3702)
| Intends to quit at W1 |
Associations between predictor variables and intending to quit smoking (n = 3702) |
|||
|---|---|---|---|---|
| Bivariate | Multivariable | |||
| n/N | % | OR (95%CI) | OR (95%CI) | |
| Number of five closest friends who use ECs | ||||
| At least one | 158/699 | 20.8 | 1.20 (0.94–1.53) | 0.99 (0.74–1.32) |
| None | 508/2704 | 18.0 | 1.00 | 1.00 |
| Frequency of seeing EC use in public | ||||
| At least some days | 439/1675 | 25.5 | 2.18 (1.77–2.67) | 1.37 (1.04–1.81) |
| All other responses | 298/2021 | 13.6 | 1.00 | 1.00 |
| General public’s attitude towards ECs | ||||
| Approves | 220/989 | 20.8 | 1.18 (0.94–1.48) | 1.13 (0.88–1.44) |
| Does not approve | 517/2710 | 18.2 | 1.00 | 1.00 |
| Number of five closest friends who smoke | ||||
| Three or more | 244/1896 | 11.7 | 0.36 (0.29–0.44) | 0.59 (0.44–0.80) |
| Two or fewer | 422/1526 | 27.1 | 1.00 | 1.00 |
| General public’s attitude towards smoking | ||||
| Approves | 58/416 | 13.6 | 0.65 (0.45–0.94) | 1.07 (0.71–1.61) |
| Does not approve | 679/3284 | 19.6 | 1.00 | 1.00 |
| Country | ||||
| Germany | 57/452 | 12.3 | 0.32 (0.23–0.46) | 0.52 (0.33–0.81) |
| Greece | 32/280 | 10.0 | 0.25 (0.16–0.41) | 0.45 (0.26–0.80) |
| Hungary | 28/251 | 11.1 | 0.28 (0.16–0.50) | 0.53 (0.27–1.02) |
| Poland | 38/351 | 8.3 | 0.21 (0.13–0.34) | 0.34 (0.18–0.65) |
| Romania | 51/364 | 13.6 | 0.36 (0.25–0.52) | 0.73 (0.44–1.20) |
| Spain | 99/614 | 14.9 | 0.40 (0.30–0.53) | 0.82 (0.54–1.25) |
| England | 432/1388 | 30.5 | 1.00 | 1.00 |
| Gender | ||||
| Female | 402/1856 | 21.3 | 1.35 (1.10–1.65) | 1.19 (0.95–1.48) |
| Male | 335/1844 | 16.8 | 1.00 | 1.00 |
| Age | ||||
| 25–39 | 189/893 | 21.7 | 1.99 (1.28–3.08) | 2.00 (1.25–3.19) |
| 40–54 | 266/1291 | 18.7 | 1.65 (1.09–2.48) | 1.62 (1.03–2.53) |
| 55 and over | 223/1205 | 18.4 | 1.61 (1.05–2.47) | 1.44 (0.89–2.32) |
| 18–24 | 59/311 | 12.2 | 1.00 | 1.00 |
| Annual income | ||||
| Low | 154/861 | 17.0 | 0.64 (0.48–0.84) | 0.79 (0.57–1.10) |
| Moderate | 246/1284 | 18.6 | 0.71 (0.55–0.92) | 0.84 (0.62–1.13) |
| Not stated | 96/662 | 14.8 | 0.54 (0.39–0.75) | 0.74 (0.51–1.09) |
| High | 241/893 | 24.3 | 1.00 | 1.00 |
| Highest level of education | ||||
| Low | 226/1342 | 15.4 | 0.61 (0.47–0.79) | 0.97 (0.71–1.33) |
| Moderate | 313/1672 | 20.1 | 0.84 (0.65–1.08) | 1.01 (0.75–1.35) |
| High | 193/659 | 23.0 | 1.00 | 1.00 |
| Use of ECs at W1 | ||||
| Current use | 263/703 | 35.5 | 2.95 (2.33–3.74) | 1.78 (1.31–2.41) |
| No current use | 474/2997 | 15.7 | 1.00 | 1.00 |
| Heaviness of Smoking Indexa | 0.83 (0.78–0.89) | 0.89 (0.82–0.95) | ||
Percentages are based on weighted data. Analysis includes all respondents who completed surveys at both waves and reported that they had heard of ECs at Wave 1. Significant results are shown in bold.
aMean (SD) = 2.16 (1.52) for those intending to quit, 2.57 (1.55) for those not intending to quit.
Prospective associations between EC social norms and attempts to quit smoking
The final set of analyses examined the prospective association between social norms and having made a quit attempt between waves (table 4). There was no association in multivariable analyses between any EC or smoking social norm variable and making a quit attempt. Attempting to quit smoking between waves was predicted by country (P = .009), intention to quit smoking (P < .001) and use of ECs at Wave 1 (P = .036).
Table 4.
Prospective associations between descriptive and injunctive EC social norms at Wave 1 and attempts to quit smoking at Wave 2 (n = 3702)
| Makes an attempt to quit between waves |
Associations between predictor variables and making an attempt to quit between Waves (n = 3702) |
|||
|---|---|---|---|---|
| Bivariate | Multivariable | |||
| n/N | % | OR (95%CI) | OR (95%CI) | |
| Number of close friends who use ECs | ||||
| At least one | 212/700 | 30.9 | 1.19 (0.93–1.54) | 1.10 (0.83–1.46) |
| None | 752/2703 | 27.3 | 1.00 | 1.00 |
| Frequency of seeing EC use in public | ||||
| At least some days | 507/1676 | 30.1 | 1.20 (0.98–1.46) | 0.94 (0.71–1.24) |
| All other responses | 535/2021 | 26.5 | 1.00 | 1.00 |
| General public’s attitude towards ECs | ||||
| Approves | 284/989 | 27.5 | 0.95 (0.76–1.19) | 0.86 (0.67–1.11) |
| Does not approve | 761/2711 | 28.5 | 1.00 | 1.00 |
| Number of close friends who smoke | ||||
| Three or more | 479/1895 | 24.9 | 0.70 (0.57–0.85) | 0.97 (0.75–1.25) |
| Two or fewer | 489/1527 | 32.2 | 1.00 | 1.00 |
| General public’s attitude towards smoking | ||||
| Approves | 104/416 | 24.4 | 0.80 (0.58–1.11) | 0.91 (0.64–1.28) |
| Does not approve | 941/3285 | 28.7 | 1.00 | 1.00 |
| Country | ||||
| Germany | 91/452 | 18.5 | 0.44 (0.32–0.61) | 0.57 (0.38–0.84) |
| Greece | 64/280 | 25.2 | 0.66 (0.45–0.98) | 0.98 (0.61–1.59) |
| Hungary | 51/251 | 18.1 | 0.43 (0.29–0.65) | 0.72 (0.44–1.17) |
| Poland | 85/351 | 27.5 | 0.74 (0.52–1.07) | 1.01 (0.63–1.61) |
| Romania | 121/364 | 29.5 | 0.82 (0.59–1.13) | 1.13 (0.73–1.73) |
| Spain | 175/613 | 27.9 | 0.76 (0.58–0.99) | 1.07 (0.72–1.60) |
| England | 458/1390 | 33.8 | 1.00 | 1.00 |
| Gender | ||||
| Female | 545/1857 | 31.1 | 1.28 (1.06–1.56) | 1.20 (0.97–1.48) |
| Male | 500/1844 | 26.0 | 1.00 | 1.00 |
| Age | ||||
| 25–39 | 262/894 | 30.6 | 0.84 (0.58–1.22) | 0.81 (0.55–1.21) |
| 40–54 | 336/1291 | 25.2 | 0.64 (0.45–0.92) | 0.73 (0.50–1.07) |
| 55 and over | 335/1206 | 25.7 | 0.66 (0.46–0.95) | 0.74 (0.50–1.09) |
| 18–24 | 112/310 | 34.4 | 1.00 | 1.00 |
| Annual income | ||||
| Low | 226/861 | 27.9 | 0.92 (0.69–1.23) | 1.02 (0.73–1.43) |
| Moderate | 366/1284 | 29.5 | 0.99 (0.77–1.29) | 1.15 (0.86–1.55) |
| Not stated | 169/662 | 24.4 | 0.76 (0.56–1.04) | 0.83 (0.58–1.18) |
| High | 284/894 | 29.7 | 1.00 | 1.00 |
| Highest level of education | ||||
| Low | 340/1342 | 23.7 | 0.61 (0.45–0.83) | 0.83 (0.58–1.18) |
| Moderate | 483/1671 | 29.8 | 0.84 (0.63–1.12) | 0.89 (0.65–1.22) |
| High | 217/661 | 33.6 | 1.00 | 1.00 |
| Intention to quit smoking at W1 | ||||
| Intends to quit | 356/737 | 46.3 | 2.75 (2.17–3.49) | 2.29 (1.77–2.94) |
| Does not intend to quit | 688/2962 | 23.8 | 1.00 | 1.00 |
| Use of ECs at W1 | ||||
| Current use | 218/479 | 42.4 | 1.84 (1.43–2.38) | 1.45 (1.03–2.05) |
| No current use | 827/2395 | 26.3 | 1.00 | 1.00 |
| Heaviness of Smoking Index a | 0.94 (0.87–1.00) | |||
Percentages are based on weighted data. Analysis includes all respondents who completed surveys at both waves and reported that they had heard of ECs at Wave 1. Significant results are shown in bold.
aMean (SD) = 2.30 (1.55) for those who made an attempt to quit between waves, 2.56 (1.54) for those who did not.
Discussion
The current study examined the associations between descriptive and injunctive social norms for EC use and smoking with initiation of EC use, the intention to quit and attempts to quit smoking. Only two out of five social norms, both descriptive, were associated with outcome variables in multivariable analyses. Having three or more friends who smoke was associated with reduced odds of intending to quit smoking and the frequency of seeing EC use in public was associated with increased odds of intending to quit, both in cross-sectional analyses. The frequency of seeing EC use in public was also associated with increased odds of initiating EC use between waves. None of the social norm variables examined here predicted whether participants made an attempt to quit smoking between waves.
There are different pathways through which seeing people use ECs in public could increase the likelihood of initiating use. Seeing others use ECs could provide cues or opportunities for EC use to take place (e.g. 9). Alternatively, it could increase susceptibility or intention to use ECs19,27 or motivate use by increasing perceptions of reduced harm relative to cigarettes.28 Based on these analyses, it does not appear that increased visibility of vaping is ‘renormalizing’ tobacco use amongst smokers and reducing their motivation to quit21,22 as seeing EC use in public was associated greater odds of intending to quit smoking. This result differs from that of a recent study23 using data collected from smokers in England which found no association between regular exposure to EC use and motivation to quit smoking. The different findings could be due to the difference in how exposure to EC use was measured; the previous study23 looked at the impact of reporting that people used ECs around them ‘regularly’, whereas the current study examined the impact of frequency of seeing people vaping in public ‘at least some days’. However, in either study, there was no support for the hypothesis that exposure to vaping decreased the intention to quit smoking. Rather, being exposed to smoking, in the form of having friends who smoke, was significantly, negatively associated with the intention to quit, as were descriptive social norms for smoking in the previous study.23 We found no evidence that injunctive social norms, either for smoking tobacco or for EC use, were associated with the initiation of EC use. In general, this is in keeping with the evidence base as findings on the association between injunctive norms and smoking behaviour or EC use have tended to be mixed (e.g. 17,18).
Descriptive EC social norms also did not impact negatively on attempts to quit smoking, although on this occasion, there was also no positive association. Rather, we found no evidence that any social norm variable, either for EC use or for smoking, was associated with actual attempts to quit. One potential explanation for the difference in the association between cognitive social norm variables and the initiation of EC use compared with attempts to quit smoking is theoretical; specifically that the uptake of novel behaviours is more readily predicted by cognitive variables such as social norms than the cessation of behaviours that have been performed regularly over time.29
This study had several limitations. The analyses examining associations between social norms and the intention to quit smoking were based on cross-sectional data meaning we cannot infer causality. Associations between EC social norms and the two behavioural outcomes: initiation of EC use and attempts to quit smoking, were, however, strengthened by the use of prospective data. Although attempts were made to control for numerous potential covariates, some important variables may have been omitted, e.g. we did not control for whether respondents had made prior attempts to quit in the analysis of whether social norms were associated prospectively with having made a quit attempt between waves. A further limitation is the use of self-report data, which leaves open the possibility that some of these findings, in particular those using cross-sectional data, could be attributed to common method bias.30 The wording of some social norms items differed marginally between the ITC 6E and ITC 4CV surveys, and differences in norms between England and the countries contributing to the ITC 6E have been reported.31 This may have impacted negatively on the internal consistency of the variables used in analyses. As has been noted elsewhere,32 the exclusion of never smokers from the current study limits the generalizability of the associations between EC social norms and the initiation of EC use. Future studies should seek to examine these associations in those who have never smoked. Finally, the power to detect differences for some comparisons may have been limited by small cell sizes.
Conclusions
This study has two main conclusions. First, as would be predicted by theoretical models of behaviour, the initiation of EC use is predicted by descriptive social norms for EC use, in particular seeing EC use in public at least some of the time. The second main conclusion is that although social norms for smoking, in terms of having three or more close friends who smoke, are negatively associated with the intention to quit, seeing EC use in public is positively associated with the intention to quit smoking. These findings may allay concerns that increasing visibility and use of ECs in society is renormalizing tobacco smoking amongst current smokers.
Key points
The initiation of EC use is predicted by descriptive social norms for EC use, namely seeing EC use in public at least some of the time.
Social norms for smoking, in terms of having three or more close friends who smoke, are negatively associated with the intention to quit, whereas seeing EC use in public is positively associated with the intention to quit.
These findings should allay concerns that increasing visibility and use of ECs in society is renormalizing tobacco smoking amongst current smokers.
Acknowledgements
EUREST-PLUS Consortium
European Network on Smoking and Tobacco Prevention (ENSP), Belgium: Constantine I. Vardavas, Andrea Glahn, Christina N. Kyriakos, Dominick Nguyen, Katerina Nikitara, Cornel Radu-Loghin and Polina Starchenko. University of Crete (UOC), Greece: Aristidis Tsatsakis, Charis Girvalaki, Chryssi Igoumenaki, Sophia Papadakis, Aikaterini Papathanasaki, Manolis Tzatzarakis and Alexander I. Vardavas. Kantar Public, Belgium: Nicolas Bécuwe, Lavinia Deaconu, Sophie Goudet, Christopher Hanley and Oscar Rivière. Smoking or Health Hungarian Foundation (SHHF), Hungary: Tibor Demjén, Judit Kiss and Anna Piroska Kovacs. Tobacco Control Unit, Catalan Institute of Oncology (ICO) and Bellvitge Biomedical Research Institute (IDIBELL), Catalonia: Esteve Fernández, Yolanda Castellano, Marcela Fu, Sarah O. Nogueira and Olena Tigova. Kings College London (KCL), United Kingdom: Ann McNeill, Katherine East and Sara C. Hitchman. Cancer Prevention Unit and WHO Collaborating Centre for Tobacco Control, German Cancer Research Center (DKFZ), Germany: Ute Mons and Sarah Kahnert. National and Kapodistrian University of Athens (UoA), Greece: Yannis Tountas, Panagiotis Behrakis, Filippos T. Filippidis, Christina Gratziou, Paraskevi Katsaounou, Theodosia Peleki, Ioanna Petroulia and Chara Tzavara. Aer Pur Romania, Romania: Antigona Carmen Trofor, Marius Eremia, Lucia Lotrean, Florin Mihaltan. European Respiratory Society (ERS), Switzerland: Gernot Rohde, Tamaki Asano, Claudia Cichon, Amy Far, Céline Genton, Melanie Jessner, Linnea Hedman, Christer Janson, Ann Lindberg, Beth Maguire, Sofia Ravara, Valérie Vaccaro and Brian Ward. Maastricht University, the Netherlands: Marc Willemsen, Hein de Vries, Karin Hummel and Gera E. Nagelhout. Health Promotion Foundation (HPF), Poland: Witold A. Zatoński, Aleksandra Herbeć, Kinga Janik-Koncewicz, Krzysztof Przewoźniak and Mateusz Zatoński. University of Waterloo (UW), Canada: Geoffrey T. Fong, Thomas K. Agar, Pete Driezen, Shannon Gravely, Anne C. K. Quah and Mary E. Thompson.
Funding
The EUREST-PLUS project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 681109 (CIV) and the University of Waterloo (GTF). Additional support was provided to the University of Waterloo by a foundation grant from the Canadian Institutes of Health Research (FDN-148477). G.T.F. was supported by a Senior Investigator Grant from the Ontario Institute for Cancer Research. E.F. is partly supported by Ministry of Universities and Research, Government of Catalonia (2017SGR319) and by the Instituto Carlos III and co-funded by the European Regional Development Fund (FEDER) (INT16/00211 and INT17/00103), Government of Spain. E.F. thanks CERCA Programme Generalitat de Catalunya for the institutional support to IDIBELL. The England arm Survey of the ITC 4 Country Smoking and Vaping Survey was supported by grants from the US National Cancer Institute (P01 CA200512) and the Canadian Institutes of Health Research (FDN-148477).
Conflicts of interest: The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. G.T.F. has served as an expert witness on behalf of governments in litigation involving the tobacco industry. A.M. is a UK National Institute for Health Research (NIHR) Senior Investigator. The views expressed in this article are those of the authors and not necessarily those of the NIHR, or the UK Department of Health and Social Care.
Contributor Information
the EUREST-PLUS Consortium:
Constantine I Vardavas, Andrea Glahn, Christina N Kyriakos, Dominick Nguyen, Katerina Nikitara, Cornel Radu-Loghin, Polina Starchenko, Aristidis Tsatsakis, Charis Girvalaki, Chryssi Igoumenaki, Sophia Papadakis, Aikaterini Papathanasaki, Manolis Tzatzarakis, Alexander I Vardavas, Nicolas Bécuwe, Lavinia Deaconu, Sophie Goudet, Christopher Hanley, Oscar Rivière, Tibor Demjén, Judit Kiss, Anna Piroska Kovacs, Esteve Fernández, Yolanda Castellano, Marcela Fu, Sarah O Nogueira, Olena Tigova, Ann McNeill, Katherine East, Sara C Hitchman, Ute Mons, Sarah Kahnert, Yannis Tountas, Panagiotis Behrakis, Filippos T Filippidis, Christina Gratziou, Paraskevi Katsaounou, Theodosia Peleki, Ioanna Petroulia, Chara Tzavara, Antigona Carmen Trofor, Marius Eremia, Lucia Lotrean, Florin Mihaltan, Gernot Rohde, Tamaki Asano, Claudia Cichon, Amy Far, Céline Genton, Melanie Jessner, Linnea Hedman, Christer Janson, Ann Lindberg, Beth Maguire, Sofia Ravara, Valérie Vaccaro, Brian Ward, Marc Willemsen, Hein de Vries, Karin Hummel, Gera E Nagelhout, Witold A Zatoński, Aleksandra Herbeć, Kinga Janik-Koncewicz, Krzysztof Przewoźniak, Mateusz Zatoński, Geoffrey T Fong, Thomas K Agar, Pete Driezen, Shannon Gravely, Anne C K Quah, and Mary E Thompson
References
- 1. Filippidis FT, Laverty AA, Gerovasili V, Vardavas CI. Two-year trends and predictors of e-cigarette use in 27 European Union member states. Tob Control 2017;26:98–104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.European Commission. Special Eurobarometer 458 “Attitudes of Europeans towards tobacco and electronic cigarettes”, 2017. Available at: http://data.europa.eu/euodp/en/data/dataset/S2146_87_1_458_ENG (11 December 2018, date last accessed).
- 3.Royal College of Physicians. Nicotine without smoke: Tobacco harm reduction. London: RCP, 2016.
- 4. Lee C, Yong H-H, Borland R, et al. Acceptance and patterns of personal vaporizer use in Australia and the United Kingdom: results from the International Tobacco Control survey. Drug Alcohol Depend 2018;185:142–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.National Academies of Sciences, Engineering, and Medicine. Public health consequences of e-cigarettes. Washington DC: National Academies Press, 2018. [PubMed]
- 6. Grana R, Benowitz N, Glantz SA. E-cigarettes: a scientific review. Circulation 2014;129:1972–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Ajzen I. The theory of planned behavior. Organ Behav Hum Decis Processes 1991;50:179–211. [Google Scholar]
- 8. Rivis A, Sheeran P. Descriptive norms as an additional predictor in the theory of planned behaviour: a meta-analysis. Curr Psychol 2003;22:218–33. [Google Scholar]
- 9. Shiffman S. Dynamic influences on smoking relapse process. J Personality 2005;73:1715–48. [DOI] [PubMed] [Google Scholar]
- 10. Ellickson PL, Bird CE, Orlando M, et al. Social context and adolescent health behavior: does school-level smoking prevalence affect students’ subsequent smoking behavior? J Health Soc Behav 2003;44:525. [PubMed] [Google Scholar]
- 11. Higgins A, Conner M. Understanding adolescent smoking: the role of the Theory of Planned Behaviour and implementation intentions. Psychol Health Med 2003;8:173–86. [Google Scholar]
- 12. Rise J, Kovac V, Kraft P, Moan IS. Predicting the intention to quit smoking and quitting behaviour: extending the theory of planned behaviour. Br J Health Psychol 2008;13:291–310. [DOI] [PubMed] [Google Scholar]
- 13. Van De Ven MO, Engels RC, Otten R, Van Den Eijnden RJ. A longitudinal test of the theory of planned behavior predicting smoking onset among asthmatic and non-asthmatic adolescents. J Behav Med 2007;30:435–45. [DOI] [PubMed] [Google Scholar]
- 14. van den Putte B, Yzer M, Willemsen MC, de Bruijn G-J. The effects of smoking self-identity and quitting self-identity on attempts to quit smoking. Health Psychol 2009;28:535. [DOI] [PubMed] [Google Scholar]
- 15. Wellman RJ, Dugas EN, Dutczak H, et al. Predictors of the onset of cigarette smoking: a systematic review of longitudinal population-based studies in youth. Am J Prev Med 2016;51:767–78. [DOI] [PubMed] [Google Scholar]
- 16. Hitchman SC, Fong GT, Zanna MP, et al. The relation between number of smoking friends, and quit intentions, attempts, and success: findings from the International Tobacco Control (ITC) four country survey. Psychol Addict Behav 2014;28:1144–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. East K, Hitchman SC, Bakolis I, et al. The association between smoking and electronic cigarette use in a cohort of young people. J Adolesc Health 2018;62:539–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Urman R, McConnell R, Unger JB, et al. Electronic Cigarette and Cigarette Social Environments and Ever Use of Each Product: A Prospective Study of Young Adults in Southern California. Nicotine Tob Res, 2019;21:1347–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Ma BH, Yong HH, Borland R, et al. Factors associated with future intentions to use personal vaporisers among those with some experience of vaping. Drug Alcohol Rev 2018;37:216–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Morello P, Pérez A, Peña L, et al. Prevalence and predictors of e-cigarette trial among adolescents in Argentina. Tob Prev Cessation 2016;2:80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Fairchild AL, Bayer R, Colgrove J. The renormalization of smoking? E-cigarettes and the tobacco “endgame”. N Engl J Med 2014;370:293–5. [DOI] [PubMed] [Google Scholar]
- 22. Voigt K, Smoking norms and the regulation of e-cigarettes. Am J Public Health 2015;105:1967–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Jackson SE, Beard E, Michie S, et al. Are smokers who are regularly exposed to e-cigarette use by others more or less motivated to stop or to make a quit attempt? A cross-sectional and longitudinal survey. BMC Med 2018;16:206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Vardavas CI, Becuwe N, Demjen T, et al. Study Protocol of European Regulatory Science on Tobacco (EUREST-PLUS): policy implementation to reduce lung disease. Tob Induc Dis 2018;16(Suppl 2):A2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Thompson ME, Fong GT, Boudreau C, et al. Methods of the ITC four country smoking and vaping survey, wave 1 (2016). Addiction 2019;114:6–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Heatherton TF, Kozlowski LT, Frecker RC, et al. Measuring the heaviness of smoking: using self‐reported time to the first cigarette of the day and number of cigarettes smoked per day. Br J Addict 1989;84:791–800. [DOI] [PubMed] [Google Scholar]
- 27. Lozano P, Arillo-Santillán E, Barrientos-Gutíerrez I, et al. E-cigarette social norms and risk perceptions among susceptible adolescents in a country that bans e-cigarettes. Health Educ Behav 2019;46:275–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. East K, Brose LS, McNeill A, et al. Harm perceptions of electronic cigarettes and nicotine: a nationally representative cross-sectional survey of young people in Great Britain. Drug Alcohol Depend 2018;192:257–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Weinstein ND. Misleading tests of health behavior theories. Ann Behav Med 2007;33:1–10. [DOI] [PubMed] [Google Scholar]
- 30. McDermott MS, Sharma R, Evaluating the impact of method bias in health behaviour research: a meta-analytic examination of studies utilising the theories of reasoned action and planned behaviour. Health Psychol Rev 2017; 11:358–73. [DOI] [PubMed] [Google Scholar]
- 31. East KA, Hitchman SC, McDermott M, et al. Social norms towards smoking and electronic cigarettes among adult smokers in seven European Countries: findings from the EUREST-PLUS ITC Europe Surveys. Tob Induc Dis 2018;16(Suppl 2):A15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Aleyan S, East K, McNeill A, Cummings KM, et al. Differences in norms towards the use of nicotine vaping products among adult smokers, former smokers and nicotine vaping product users: cross‐sectional findings from the 2016 ITC Four Country Smoking and Vaping Survey. Addiction 2019;114:97–106. [DOI] [PMC free article] [PubMed] [Google Scholar]