Scottish Health Survey 2014 - volume 1: main report

3 SMOKING

Linsay Gray & Alastair H Leyland

SUMMARY

Smoking prevalence

• In 2014 one in five adults (22%) reported that they were current cigarette smokers, similar to the rate in 2013 (21%).
• Just under a quarter (23%) of adults used to smoke regularly and 55% had either never smoked at all, or used to smoke but not regularly.
• Prevalence of current smoking in 2014 was similar among men (23%) and women (21%).
• Current smoking was higher among those aged 25-64 (23-26%) than those aged 16-24 (20%), 65-74 (16%) and 75 and over (10%).
• Adult smokers in 2014 smoked on average 13.2 cigarettes per day. Younger smokers smoked the least cigarettes on average per day (10.8 for those aged 16-24 and 10.3 for those aged 25-34).

Children's exposure to tobacco smoke in the home

• The proportion of children reported to be exposed to second-hand smoke in the home remained at 11% in 2014 (the same figure as in 2013).
• Exposure levels increased with age from 4% for those aged 0-1 to 18% for those aged 13-15.

Adults' use of e-cigarettes

• In 2014, 15% of adults aged 16 and over reported ever having used an e-cigarette, including 5% who said they currently used them. Prevalence of ever using was highest among the younger age groups while current usage was highest for those aged 35-64 (6-7%).
• Exactly half (50%) of current cigarette smokers reported ever using e-cigarettes, including 15% currently using. In total, 14% of ex-regular smokers ever used an e-cigarette including 7% currently using them. Just 1% of people who had never smoked regularly reported using e-cigarettes currently or ever.

Quit attempts and smoking cessation

• In 2014, two-thirds (67%) of smokers said they would like to quit smoking.
• Frequency of attempts to quit showed 21% of all smokers having made no attempts, 37% one or two and 43% at least three attempts.
• Women were more likely than men to have attempted to quit smoking. In total, 82% of female smokers had attempted to quit, compared with 77% of male smokers. Just under half (46%) of female smokers, and 39% of male smokers, had attempted to quit on at least three occasions.
• Patterns by age showed a link between the length of time someone has smoked and their number of quit attempts, with smokers from the age of 35 onwards the most likely to have made three or more unsuccessful quit attempts.
• Just under two-thirds (64%) of recent ex-smokers and current smokers who had attempted to quit said they used a nicotine replacement therapy (NRT) product or e-cigarettes in the previous three months. Use of products to support a quit attempt was significantly higher for women (67%) than men (60%), and also higher among those aged 18-64 (64-68%) than those aged 65 and over (50%).
• The most common items used as part of a recent quit attempt were nicotine patches (36%) and e-cigarettes (32%).
• Over a third (36%) of people who had used a product as part of a recent quit attempt had also used smoking cessation services, most commonly pharmacies (18%), specialist cessation advisers (9%) and GPs (7%).

Factors associated with successful quitting

• Factors associated with having successfully quit smoking for adults who had ever smoked regularly included:
  • Age - those aged 55 and over had increased odds of having successfully quit, compared with those aged 16-34.
  • Marital status - married / in civil partnership had increased odds of having quit.
  • Body mass index - obese / morbidly obese adults had increased odds compared with those not overweight.
  • SIMD - those in the most deprived areas had lower odds (women only).
  • Equivalised household income - those with lower household incomes had lower odds.
  • Economic activity - those who were unemployed and looking for work had lower odds than those in employment.
  • Education - those with no or only low-level qualifications had lower odds than those with a degree.
  • Alcohol consumption - those who drank over the weekly recommended maximum number of units had lower odds (men only).

3.1 INTRODUCTION

Nationally^[1] and globally,^[2] tobacco use is the leading cause of preventable poor health and premature mortality, each year killing around 6 million people and costing over half a trillion dollars worldwide.^[3] In Scotland alone, tobacco use is associated with over 13,000 deaths (around a quarter of all deaths) and approximately 56,000 hospital admissions every year.^[4]

3.1.1 Policy background

Several of the Scottish Government's National Indicators are relevant to smoking.^[5] In addition to the specific indicator to reduce the proportion of adults who are current smokers, there are more general related indicators on, for example, reducing premature mortality and reducing emergency admissions to hospital.^[6]

The Tobacco Control Strategy^[7] lays out the Scottish Government's vision to create a 'tobacco-free generation' (defined as 'a smoking prevalence among the adult population of 5% or lower') by the year 2034. Actions arising from the strategy are structured around the themes of prevention, protection and cessation. Smoking cessation interventions, including pharmacotherapy, are among the most cost-effective health care interventions available.^[8]

One outcome of the actions under the cessation theme was the development of the NHS Scotland HEAT target to achieve at least 12,000 successful quits at twelve weeks post quit, in the 40% most deprived areas within each NHS Health Board (60% for Island Boards) over the one year ending March 2015.^[9] These are being replaced in 2015 with the new NHS Local Delivery Plan (LDP) Standards which require NHS Boards to sustain and embed successful smoking quits at twelve weeks post quit, in the 40% most deprived SIMD areas (60% in the Island Boards).^[10]

3.1.2 Reporting on smoking in the Scottish Health Survey (SHeS)

Reliable data on smoking behaviour, cessation, NRT use and exposure to second-hand smoke are vital to effective monitoring of trends relevant to the various targets in place. The SHeS self-reported data presented in this chapter complement the data provided by the Scottish Household Survey which is used to measure the current NPF indicator on reducing smoking among adults.^[11] This chapter presents figures for prevalence of cigarette smoking, cessation attempts and support, NRT use, e-cigarette use and children's exposure to second-hand smoke.

Nicotine replacement therapy (NRT) products supply low doses of nicotine but do not contain the toxins found in tobacco smoke. The goal of their use is to reduce nicotine cravings and ease the symptoms of nicotine withdrawal. Nicotine replacement products come in many forms such as inhalers, gum, lozenges, nasal sprays and skin patches. From 2014, SHeS has gathered information on the use of e-cigarettes among the Scottish adult population, in response to their increased availability and high profile.

3.1.3 Comparability with other UK statistics

The Health Survey for England, Health Survey for Northern Ireland and Welsh Health Survey provide estimates of smoking prevalence in the other home nations within the UK. A Government Statistical Service publication on the comparability of official statistics across the UK advises that the smoking prevalence estimates across these surveys are only partially comparable as they are conducted separately and have different sampling methodologies.^[12] Smoking prevalence estimates from the UK-wide Integrated Household Survey for Scotland, Wales, England and Northern Ireland have been deemed as fully comparable.

3.2 METHODS AND DEFINITIONS

3.2.1 Methods of collecting data on smoking behaviour

Adults aged 20 and over were asked about their smoking behaviour during the face to face interview. For those aged 16 and 17, information was collected in a self-completion questionnaire offering more privacy and reducing the likelihood of concealing behaviour in front of other household members. At the interviewer's discretion those aged 18 and 19 could answer the questions either face to face or via the self-completion booklet.

3.2.2 Questions on smoking behaviour

Questions on smoking have been included in SHeS since 1995. Some small changes were made to the questions in 2008 and 2012, these are outlined in the relevant annual reports.^[13],^[14]

The current questions in the survey focus on:

• current smoking status
• frequency and pattern of current smoking
• the number of cigarettes smoked by current smokers
• ex-smokers' previous smoking history
• exposure to second-hand smoke
• past smoking behaviour
• quit attempts and desire to give up smoking
• medical advice on giving up smoking
• NRT use
• e-cigarette use (including as part of a quit attempt)

While the self-completion questions were largely similar to those asked in the face to face interview, the self-completion questionnaire did exclude questions on: past smoking behaviour, desire to give up smoking and medical advice to stop smoking.

3.2.3 Definitions

Cigarette smoking status

Information on cigar and pipe use is collected in the survey but as prevalence is low these are not considered in the definition of current smoking. Smoking status categories reported here are:

• current cigarette smoker
• ex-regular cigarette smoker
• never regular cigarette smoker
• never smoked cigarettes at all

Children's exposure to second-hand smoke

Children's (age 0-15) exposure to second-hand smoke is measured in two ways in the survey:

• whether there is someone who regularly smokes inside the accommodation where the child lives, and
• parents' and older children's (aged 13-15) reports of whether children are exposed to smoke at home.

3.3 CIGARETTE SMOKING STATUS

3.3.1 Trends in cigarette smoking status since 1995

The results in each of the self-reported cigarette smoking status groups for adults are given in Table 3.1 for all relevant years. As there have been changes to the age range of the sample over the years, data are presented for all adults aged 16 and over since 2003 along with data for individuals aged 16-64 from 1995.

Table 3.1 presents the trends in smoking status for all adults aged 16 and over since 2003. This showed a current smoking level of 28% in 2003 and a more recent decline between 2012 and 2013 (from 25% to 21%) with the level in 2014 at 22%. This decline from 2012 was statistically significant, providing evidence of progress on the NPF National Indicator to reduce the percentage of adults who smoke.7 As outlined in section 3.1.3, progress towards the Indicator is monitored using the Scottish Household Survey which had a smoking estimate of 20% in 2014.

While the proportion of all adults aged 16 and over who had never smoked, or had never smoked regularly, increased from 50% in 2003 to 55% in 2011, this figure has remained relatively static since then, at 54-55%. The proportion of all adults identifying as ex-regular smokers changed little between 2003 and 2014 (remaining between 22 and 24%). Time-related patterns and levels for smoking status among those aged 16 and over were similar for men and women.

Figures for all adults aged 16-64 showed a similar significant decline in current smoking levels but with this trend having begun in 1998 (35%) and continued to 2014 (24%).

Among all smokers aged 16 and over, there was also a significant fall over time in the mean number of cigarettes smoked per day from 15.3 cigarettes in 2003 to 13.2 cigarettes in 2014. Figures for adults aged 16-64 show that this decline over time can be traced further back, to 1995 (a mean of 16.7 cigarettes in 1995, 15.3 in 2003, and 13.1 in 2014). This overall decrease was more apparent among male smokers aged 16-64 (18.1 cigarettes per day in 1995 to 13.1 cigarettes in 2014) than female smokers (15.4 cigarettes and 13.1 cigarettes respectively). Figure 3A, Table 3.1

3.3.2 Cigarette smoking status in 2014

Data on self-reported cigarette smoking status for all adults aged 16 and over in 2014 are shown in Table 3.2. Just over one in five (22%) adults were current smokers in 2014, with 23% reporting that they used to smoke regularly and over half (55%) that they had either never smoked at all, or used to smoke, but not regularly. Current smoking prevalence was similar for men (23%) and women (21%); none of the other figures for smoking status varied by sex either.

There were marked variations in cigarette smoking status by age, as reported previously in SHeS.14 The prevalence of current smoking in 2014 was highest among those aged 25-64 (23% to 26%), lower among those aged 16-24 (20%), and lowest among those aged 65-74 (16%) and those aged 75 and over (10%). The lower current smoking prevalence among the youngest and oldest age groups was true for both sexes.

As would be expected, the proportion of people identifying as ex-regular smokers was lowest for the youngest age group (7% for those aged 16-24) and highest at older ages (39% for those aged 65-74, 41% for those aged 75 and over) in 2014. This correlation corresponded with patterning by age in the proportions reporting that they had never smoked or had never smoked regularly (72% for those aged 16-24 compared with 44% for those aged 65-74 and 49% for those aged 75 and over).

Table 3.2 shows the mean number of cigarettes smoked per day per adult smoker in 2014. The overall mean was 13.2 cigarettes, and was similar for male (13.5 cigarettes) and female smokers (13.0 cigarettes). In 2014, younger smokers smoked the least cigarettes on average per day (10.8 cigarettes for smokers aged 16-24, 10.3 cigarettes for those aged 25-34) with higher average numbers for those aged 45-74 (between 15.0 and 15.7 cigarettes). These age-related patterns for numbers of cigarettes consumed were similar for male and female smokers. Figure 3B, Figure 3C, Table 3.2

3.4 CHILDREN'S EXPOSURE TO SECOND-HAND SMOKE IN 2014

The two measures of children's exposure to smoke at home (described in 3.2.3) are presented for 2014 in Table 3.3. The first set of figures shows the prevalence of children living in accommodation in which someone smokes inside. The second figure is for children's reported exposure to smoke in the home, which is being used to monitor progress towards the target to reduce this to 6% by 2020.

Overall, 16% of children (17% of boys and 16% of girls) lived in accommodation in which someone smoked inside. However, a lower proportion of 11% of children (12% of boys and 10% of girls) were reported to be exposed to second-hand smoke in their home. This figure has not changed significantly from 2012 (12%) or 2013 (11%).

Reported exposure to smoke increased with age, from 4% for those aged 0-1 to 18% for those aged 13-15, with similar levels seen for both boys and girls in each age group. There was less variation by age in the proportion of children who live in accommodation in which someone smokes inside, with levels lowest for those aged 0-1 (10%) but varying between 13% and 21% for all other age groups. Figure 3D, Table 3.3

3.5 E-CIGARETTE USE IN 2014

Use of e-cigarettes in the adult population for 2014 is given by age and sex in Table 3.4. In total, 5% of adults aged 16 and over currently used e-cigarettes, with a further 10% having previously used them (15% therefore having ever used). These figures were similar for both men and women.

Use of e-cigarettes varied by age in 2014, as shown in Figure 3E and Table 3.4. The population prevalence of ever use (including those currently using) decreased steadily with advancing age from 20% among adults aged 16-24 to 3% among those aged 75 and over. The prevalence of current use of e-cigarettes in 2014 was highest among the middle age groups (6-7% among adults aged 35-64) and lower for younger and older adults (4% among those aged 16-34, 1-3% for those aged 65 and over). Figure 3E, Table 3.4

Table 3.5 shows that e-cigarette use was strongly associated with smoking behaviour, with both current and past use of e-cigarettes much higher among current cigarette smokers than among ex-regular or never regular smokers. In total, 15% of smokers reported currently using e-cigarettes and an additional 35% said they had done so in the past, the equivalent figures for ex-regular smokers were 7% and 7%, respectively. Prevalence of e-cigarette use either now or ever was 1% among people who had never smoked regularly (or at all). Use of e-cigarettes as part of smoking quit attempts is discussed in the next section. Table 3.5

3.6 QUIT ATTEMPTS AND SMOKING CESSATION

3.6.1 Quit attempts and aspirations in 2014

Table 3.6 presents 2014 data on the number of attempts current smokers said they had made to quit smoking, and the proportions who said they would like to stop smoking. A fifth (21%) of smokers had made no attempts to quit smoking, 37% had made one or two attempts, and a further 43% had made three or more attempts to quit. Two thirds of smokers (67%) said they would like to quit smoking (61% of male, and 73% of female smokers).

In 2014, a somewhat greater proportion of male than female smokers had made no attempt to quit (23% and 18%, respectively). The proportions making one or two attempts to quit were more similar (37% and 36%, respectively). However, male smokers were less likely than female smokers to have made three or more attempts to quit (39% and 46%, respectively).

Table 3.6 shows the differences in quit attempts by age in 2014. The proportion of smokers having made no quit attempts tended to decline with age from those aged 18-34 (29%) to those aged 55-64 (13%) before increasing for those aged 65 and over (23%). Younger smokers were also the least likely to have made three or more attempts to quit (30% of smokers aged 18-34 compared with 42-50% of smokers aged 35 and over). The age-related associations were generally similar for men and women, though the sample sizes are relatively small for some of the sex-specific age groups.

These patterns may reflect a correspondence between quit attempts and the total length of time someone has smoked, rather than an association between being a particular age and wanting to stop smoking, as shown by the figures on smokers' desire to quit. Smokers aged 35-64 had similar proportions wanting to quit (67-77%) with lower levels seen for the 18-34 age group (64%) and those aged 65 and over (51%). Female smokers appeared to be keener to quit than their male counterparts across all age groups, (although the results for any individual age group were not statistically significant) except for those aged 65 and over. Table 3.6

3.6.2 Products to support quit attempts in 2014

All current smokers who had ever attempted to quit, and recent ex-smokers (who had quit within the past year), were asked whether they had used any of a list of products as part of a quit attempt within the previous three months. In 2014, 64% of this group of smokers and recent ex-smokers had used some form of nicotine replacement therapy (NRT), or e-cigarettes, for this purpose in the last three months, with the figure significantly higher for women (67%) than men (60%). The particular items most likely to have been used as part of the quit attempt were nicotine patches (36%) and e-cigarettes (32%). Nicotine gum and nasal sprays / nicotine inhalers were used by 17% and 9% respectively, with other products - Lozenge / microtab, Champix / Varenicline and Zyban / Bupropion - being less common.

In 2014, use of products as part of a recent quit attempt varied significantly with age, with those aged 18-64 (64-68%) more likely than those aged 65 and over (50%) to have used at least one of these items. Most of this difference between the age groups was accounted for by patterns in e-cigarette use, which were used by 34-35% of those aged 18-64 who had made a recent quit attempt, but just 16% of those aged 65 and over. Table 3.7

3.6.3 Additional smoking cessation support in 2014

Table 3.8 displays the data for 2014 on use of various services for smoking cessation support during the previous three months. These questions were only asked of people who had used at least one of the products described above. Cessation support services were used by 36% of smokers / ex-smokers who had used a product as part of a recent quit attempt (34% of men and 38% of women in this group). Service use was similar for men and women for pharmacies (18% of all product users), specialist cessation advisors (9%), and GPs (7%). However, cessation support from a GP practice nurse was used more often by women (6%) than men (2%). Use of cessation support services was significantly associated with older age, with 51% of people aged 65 and over with a recent quit attempt using some form of cessation support in combination with a product, compared with 34-35% of those aged under 65 (though note that the sample size for the oldest group is quite small, so this estimate is not very precise). Table 3.8

3.7 FACTORS ASSOCIATED WITH SUCCESSFUL QUITTING

Multivariable logistic regression was used to examine the independent effect of a range of socio-demographic and behavioural factors associated with successful quitting of smoking (the dependent variable) among adults who had ever smoked regularly (i.e. former and current smokers). The factors investigated in the regression models shown in Table 3.9 included behavioural characteristics explored in other chapters in this report: consumption of alcohol (drinking outwith the weekly guidelines), and body mass index (BMI); as well as the key socio-demographic factors of age, marital status, economic activity, education, area deprivation (SIMD), and equivalised household income. By simultaneously controlling for a number of factors, the independent effect each factor has on the variable of interest can be established. Other factors were included in preliminary models, but excluded from the models presented on the basis that they did not show any significant independent associations with successful quitting for either men or women. These included physical activity, levels of fruit and vegetable consumption, being a parent, and socio-economic classification (NS-SEC), both of the respondent at the time of interview and of the family in which they were brought up.

Logistic regression compares the odds of a reference category (shown in the table with a value of one) with that of the other categories. Regressions were run on 2014 data with separate models for men and women. The odds ratios for adult former and current smokers having successfully quit smoking (i.e. being an ex-smoker) are presented in Table 3.9. An odds ratio of less than one means that the odds of someone from a given group having quit smoking are lower than that of an individual from the reference category. When the odds ratio is greater than one, the opposite is true. Independent variables with a p-value of 0.05 or less are significant predictors of the dependent variable at the 95% confidence level.

Age group, marital status, body mass index and SIMD were found to be significant predictors of having successfully quit for both men and women. Additionally, equivalised household income, economic activity and (albeit only just) alcohol intake were significant for men, and level of education was significant for women.

Once all other factors were controlled for, men and women aged 55 and over had significantly higher odds of having successfully quit smoking than those aged 16-34: the odds ratio for men aged 55 to 64 was 2.60, for 65 to 74 was 7.32 and for 75 and over was 25.63; the respective odds ratio for women were 1.77, 5.07 and 6.56.

Men and women who were single (odds ratios of 0.51 for men, 0.40 for women) or separated / divorced / had a dissolved civil partnership (0.29 for men, 0.60 for women) had significantly lower odds of having quit smoking than their married / civilly partnered counterparts. The relative odds for men who were living with a partner but not married / in civil partnership (0.58) were also significantly lower albeit only marginally so.

The odds of women living in the most deprived areas (odds ratio of 0.50) successfully quitting smoking were exactly half those of women in the least deprived ones. For men, while SIMD shows a significant association with having quit smoking, the individual differences between deprivation quintiles presented are not significant, although those in the most deprived quintile (0.59) had lower odds compared with all other groups (1.00 to 1.31).

Compared with those who were not overweight (BMI <25 kg/m2), obese men and women (BMI 30 kg/m2 and over) had increased odds of having quit smoking (odds ratios of 2.14 for men and 1.82 for women).

Men exceeding the recommended weekly alcohol consumption guidelines for their sex had significantly lower odds of having quit smoking compared with those consuming less than the recommended amounts (odds ratio of 0.66).

There were lower odds of having quit smoking for men living in households in the 4th and 5th (lowest) income quintile (odds ratios of 0.46 and 0.55 respectively) than those living in households in the highest income quintile, although these were only significantly lower for the 4th quintile. Additionally, the odds of women having quit smoking were significantly lower (0.50) for those in the lowest income quintile relative to those in the highest income quintile.

Men who were unemployed and looking for paid work (odds ratio of 0.22) had significantly lower odds of having quit smoking than those in paid employment, self-employed or on government training. Unemployed women also had significantly lower odds (0.40), although the overall association between economic status and having quit smoking was not significant.

Women with no qualifications (odds ratio of 0.48), with school qualifications below standard grade (0.47), or with standard grade or equivalent qualifications (0.51) all had decreased odds of having quit smoking, compared with those with a university degree or equivalent / higher. Men with no qualifications (0.54) or with school qualifications below standard grade (0.45) also had lower odds, although the overall association for men between education and having successfully quit was not significant. Table 3.9