Sample and recruitment
Data were drawn from the ongoing Smoking and Alcohol Toolkit Study (STS), a monthly repeated cross-sectional survey of a representative sample of adults (≥18 years) in Scotland, Wales and England.27
The sample consisted of respondents in Scotland, Wales and England surveyed in September 2021.
The STS uses a hybrid of random location and quota sampling to select a new sample of approximately 2,400 adults each month (~1800 in England, ~350 in Scotland and ~250 in Wales). Telephone interviews are performed with one household member until quotas based on factors influencing the probability of being at home (e.g., gender, age, working status) are fulfilled.
The telephone interviews are conducted by landline and mobile using a standard landline random digit dialling (RDD), mobile RDD, and targeted mobile. In terms of the sample processed, each eligible landline telephone number across GB has a random probability of selection proportionate to population distribution (i.e., stratification of the landline telephone database by and within Government Office Region, GOR). Within GOR, the system is based on UK postcode sector information. Each postcode sector is matched to the relevant standard telephone dialling code and telephone number stubs are derived from information obtained from the Office of Communications (Ofcom). Selection probability of postcode sectors is proportional to the number of households within or across a given area by using the household density information that is attached to each postcode sector. Mobile sampling uses largely the same approach as landline sampling; however, the selection is in proportion to the known mobile network share. This mobile network share is continually updated using robust publicly available statistics to ensure that accurate samples of the mobile using population. Mobile, targeted mobile and landline sampling are carried out in equal proportions. To maximise response rates more landline sampling takes places earlier in the day, with more mobile sampling later in the day.
Data were weighted to match the Great Britain population profile on age, social grade, region, tenure, ethnicity, and working status within sex. The dimensions are derived monthly from a combination of the English 2011 census, Office for National Statistics mid-year estimates, and an annual random probability survey conducted for the National Readership Survey.27
Measures
Support for novel tobacco availability policies
Policy ideas were selected following consultation and review with academic, government and advocacy stakeholders associated with the Shaping public health policies to reduce inequalities and harm (SPECTRUM) consortium (www.spectrum.ac.uk). Respondents were asked to indicate whether they would support the following statements presented in a random order (response options: Strongly support/Tend to support/No opinion either way/Tend to oppose/Strongly oppose/Unsure or Don’t know):
- Ban the sale of cigarettes and tobacco products to everyone born after a certain year from 2030 onwards
- Raising the legal age of sale of cigarettes and tobacco from 18 to 21
- Requiring anyone selling tobacco to have a licence which can be removed if they sell to those under-age
- Reducing the number of retailers selling cigarettes and tobacco in neighbourhoods with a high density of tobacco retailers.
- Restricting the sale of cigarettes and tobacco in close proximity to schools.
For prevalence estimates, responses of “strongly support” or “tend to support” are presented as ‘Yes’, responses of “Tend to oppose” or “Strongly oppose” are presented as ‘No’, and responses of “No opinion either way” or “Don’t know/unsure” are presented as ‘No opinion/unsure’.
For regression models, responses under ‘No’ and ‘Not sure’ were collapsed into ‘Do not support’ to create a dichotomous outcome variable.
Nation in GB
GB data were split into Scotland, Wales and England using government office region classifications.
Sociodemographic characteristics
The sociodemographic characteristics sex (categorized as women vs other (including men or ‘in another way’), age (continuous variable) and social grade (ABC1: higher and intermediate managerial, administrative and professional, supervisory, clerical and junior managerial, administrative and professional; C2DE: skilled manual workers, semi-skilled and unskilled manual workers and state pensioners, casual and lowest-grade workers, unemployed with state benefits) were included.
Whether or not respondents have children at home was derived from a question regarding household status. Responses were dichotomised into ‘Yes’ or ‘No’ indicating the presence or absence of children at home, respectively.
Whether respondents live in an urban or rural location was also measured. Using local authority code data, respondents were classified according to whether they live in rural, suburban, urban or metropolitan area. For ease of interpretation, suburban, urban and metropolitan will be collapsed into ‘urban’.
Smoking status
Smoking status was ascertained using the following question and response options:
“Which of the following best applies to you?”
1. I smoke cigarettes (including hand rolled) every day
2. I smoke cigarettes (including hand rolled), but not every day
3. I do not smoke cigarettes at all, but I do smoke tobacco of some kind (e.g. Pipe, Cigar or Shisha)
4. I have stopped smoking completely in the last year
5. I stopped smoking completely more than a year ago
6. I have never been a smoker (ie. smoked for a year or more)
Responses of 1 or 2 above were classified as current cigarette smokers, 4 or 5 as ex-smokers and 6 as never smokers.
Those who indicate that they do not smoke cigarettes but do smoke tobacco of some kind (answer 3 above, N=33) were excluded from the analysis because they do not include measures that assess dependence in cigarette smokers (cigarettes per day and time to first cigarette after waking).
Cigarette dependence
Cigarette dependence was measured using a measure of strength of urges to smoke (SUTS).28 SUTS has been found to be a useful measure of cigarette dependence28 and is based on responses to two questions:
‘How much of the time have you felt the urge to smoke in the past 24 hours?
- ‘Not at all’
- ‘A little of the time’
- ‘Some of the time’
- ‘A lot of the time’
- ‘Almost all of the time’
- ‘All of the time’
- ‘Don’t know’
‘In general, how strong have the urges to smoke been?’
- ‘Slight’
- ‘Moderate’
- ‘Strong’
- ‘Very strong’
- ‘Extremely strong’.
- ‘Don’t know’
Those who respond that they experienced the urge to smoke ‘Not at all’ in the first question were coded as 0 to create a final strength of urges scale ranging from 0 (no urges) to 5 (extremely strong urges). Those who respond with ‘Don’t know’ to either question were excluded.
The heaviness of smoking index (HSI)29 was used as an alternate measure of dependence in a sensitivity analysis. The HSI uses two questions from the Fagerström Test for Nicotine Dependence: time to first cigarette in the morning after waking and the number of cigarettes smoked per day. Those with a score >4 are considered to have high dependence, and those with <4 considered to have low/moderate dependence.30
Motivation to stop smoking
Motivation to stop smoking was assessed using the Motivation To Stop Scale,31 a single-item measure with seven response options representing increasing motivation to quit:
Motivation to quit
- ‘I don’t want to stop smoking’
- ‘I think I should stop smoking but don’t really want to’
- ‘I want to stop smoking but haven’t thought about when’
- ‘I REALLY want to stop smoking but I don’t know when I will’
- ‘I want to stop smoking and hope to soon’
- ‘I REALLY want to stop smoking and intend to in the next 3 months’
- ‘I REALLY want to stop smoking and intend to in the next month’.
For ease of interpretation, responses were collapsed into two variables reflecting high (6–7) vs. low (1–5) motivation to stop smoking.32
Past year quit attempt
Quit attempts in the past-year were measured among past year smokers using the question “How many serious attempts to stop smoking have you made in the last 12 months?”
We distinguished those who had not attempted to quit in the last year versus those who made 1 or more attempts.
Analysis
Respondents with missing data on any of the variables of interest were excluded from the analyses (less than 5% of responses). Characteristics of the sample and levels of support overall and within each GB nation are presented using weighted descriptive statistics.
Research question 1
Weighted prevalence data on support for each policy option are presented for GB overall, and for descriptive comparisons between Scotland, Wales and England.
Research question 2
All variables (age, sex, social grade, children in the household, urban/rural location, smoking status) were included in multivariable logistic regression models with the whole sample to evaluate which, if any, of the assessed sociodemographic and smoking status variables are independently associated with favouring each tobacco availability policy option, respectively. Models were constructed for both the overall Great Britain sample and also stratified by nation to provide within country estimates.
Research question 3
Similar multivariable logistic regression models were constructed with the sample restricted to past-year smokers to evaluate which, if any, of the smoking and quitting characteristics (past year quit attempts, motivation to stop, SUTS) are associated with favouring each tobacco availability policy option, respectively. Due to the reduced sample size, models were constructed for the overall Great Britain sample only (see below).
Sensitivity analyses
The same analyses as in research question 3 were conducted with cigarette dependence measured using HSI instead of SUTS.
Analyses were pre-registered on the open science framework (https://osf.io/mtwxe) carried out in R studio using R version 4.0.3.
Unregistered changes to analysis plan
Sparse data precluded stratified nation analyses when selecting the past-year smoker samples. Analyses were therefore only conducted using the overall Great Britain sample. An additional sensitivity analysis was run excluding individuals reporting “Don’t know/unsure” in response to each respective policy support question.