The Scottish Health Survey 2008

4 SMOKING

Linsay Gray and Alastair Leyland

SUMMARY

• 27% of men and 25% of women aged 16 and over reported that they smoked cigarettes in 2008. This varied with age: from 28% in men aged 16-24 rising to a peak of 36% in those aged 25-34 before falling consistently with increasing age to 10% of men aged 75 and over. For women, smoking prevalence was around 29% for those aged 16 to 54, then decreased with age to 11% of women aged 75 and over.
• The prevalence of cigarette smoking in adults aged 16 to 64 has decreased from 34% in 1995 to 29% in 2008 for men, and from 36% to 28% for women.
• Non-smokers' exposure to second-hand smoke has declined markedly in the decade since 1998. Exposure to second-hand smoke in the participants' own home has decreased from 18% to 10% in both men and women, while exposure in other people's homes declined from 21% to 12% among men, and from 25% to 13% among women.
• With the introduction of the ban on smoking in public places, exposure in pubs and on public transport has been virtually eliminated. In 1998 44% of men and 30% of women said they were exposed to smoke in pubs and 7% of men and 8% of women experienced this on pubic transport. In 2008 just 1% of men and less than 1% of women reported being exposed to smoke in these places.
• Of current smokers aged 16 and over, 34% of men and 28% of women smoked heavily. Figures were higher for those aged 35 to 74.
• The mean number of cigarettes smoked per day by current smokers aged 16 or over was 15.7 cigarettes for men and 13.7 cigarettes for women.
• The age-standardised prevalence of smoking varied between different socio-economic classifications. 36% of men in semi-routine and routine households were current smokers compared with 17% in professional and managerial households. The equivalent figures for women were 38% and 17%, respectively.
• Similar patterns were found when looking at household income and area deprivation. 44% of men and 39% of women in the lowest household income quintile were current smokers compared with 18% of men and 16% of women in the highest income quintile. The age-standardised prevalence of smoking in the most deprived quintile was 39% for men and women compared with 15% of men and 14% of women in the least deprived quintile.
• Logistic regression demonstrated that age, socio-economic classification, equivalised household income, area deprivation, marital status, body mass index, fruit and vegetable consumption and alcohol intake were all significantly associated with smoking prevalence.
• 34% of men and women current smokers aged 18 or over had received medical advice to give up smoking at some point in time, 14% had received this advice in the previous year.
• Of those aged 18 or over who had stopped smoking regularly, 4% of men and 7% of women had done so in the six months prior to the survey interview. The figures extended to 9% of men and 12% of women when considering those who had stopped within the previous 2 years.

4.1 INTRODUCTION

Smoking is the single biggest cause of preventable premature death and ill-health in Scotland and is implicated as a factor in all of Scotland's biggest killer diseases of cancer, coronary heart disease and stroke. Approximately 13,500 people die every year from smoking-related illness in Scotland, of which an estimated 1,500-2000 are due to second-hand smoke. ^{1, 2, 3} Many of these smoking-related deaths are avoidable deaths, with the average years of life gained by smokers who quit ranging from 10 years for those stopping at 30 to three years for those stopping at 60. ⁴ It is estimated that eradication of smoking would eliminate one-third of cancer deaths and one-sixth of deaths from other causes, including cardiovascular and chronic respiratory diseases. ⁵ In addition, the financial burden of smoking on Scotland's economy as a whole has been estimated at £837 million per year, including the costs associated with treating ill-health, lost employee productivity and reduced consumer spending as a consequence of early death. ⁶ Smoking levels and related mortality are strongly socially patterned, being highest among more socially disadvantaged groups. ⁷ Consequently, reducing the effects and impacts of smoking-related harm is a major component of the Scottish Government's agenda towards addressing health improvement and health inequalities.

In 2008 the Scottish Government published its Action Plan Scotland's Future is Smoke-Free. ⁸ This document includes a whole raft of actions and measures that are to be undertaken in the coming years, and reports that an additional £9m over the 3 years 2008/09 to 2010/11 is being made available to support action outlined in the Plan, bringing the total dedicated resources for tobacco control to £42m over the same period. With the support of this additional funding, action is going to be taken to restrict the display of cigarettes and other tobacco products at the point of sale and update statutory controls on the sale of tobacco products.

The Action Plan set out a vision of smoking reduction for Scotland in 2020 and builds on a number of major developments relating to tobacco-control in the past decade. These include the comprehensive programmes of action set out in the 1998 UK White Paper Smoking Kills, ⁹ the first ever action plan on tobacco designed specifically for Scotland - A Breath of Fresh Air for Scotland (2004), ¹⁰ and the 2006 report of the Scottish Smoking Prevention Working Group Towards a Future Without Tobacco. ¹¹ A great deal of funding has been invested in tobacco control activities, which has helped create a national network of smoking cessation services across Scotland. The Action Plan also argues that major inroads have been made in shifting cultural attitudes to smoking through effective multi-media communications campaigns and firm legislative action.

There has been significant legislative change since the publication of the last Scottish Health Survey report in autumn 2005, the most significant of which was the introduction of the ban on smoking in public places introduced in March 2006; additionally, the age at which it is legal to purchase tobacco was increased, from 16 years to 18 years, in October 2007. Analysis carried out in advance of the ban in public places estimated that comprehensive implementation of the Scottish legislation may ultimately lead to a reduction of over 400 deaths a year plus other significant health gains. ¹² Research carried out after the ban suggests there have been improvements in the respiratory health of bar workers, while another study showed a sharp increase in the rate of decline for hospital admissions for heart attacks in the six months immediately post-ban. ^13,2,14 Evidence from longer term evaluations will be needed before firm conclusions can be drawn about its impact.

One of the Scottish Government's National Performance Framework national indicators, ¹⁵ as set out in Chapter 1 of this Report, relates directly to smoking:

Reduce the percentage of the adult population who smoke to 22% by 2010

The target set out in the national indicator has its origins in the targets published in the 1999 White Paper Towards a Healthier Scotland. ¹⁶ At that time the aim was for smoking rates among those aged 16-64 to decline from 35% in 1995 to 33% in 2005, and to 31% by 2010. The latter figure was reduced in an update in 2004, contained in A Breath of Fresh Air for Scotland, ¹⁰ to 29% by 2010. In January 2006 the target was expanded to cover the whole adult population and was re-set to 22%. This latest target is reiterated within the 2008 Action Plan:

To reduce smoking among adults (16 and over) from 26.5% in 2004 to 22% by 2010

To reflect the fact that a number of the actions within the Plan have a specific emphasis on young people an additional target has now been set for those aged 16-24:

To reduce the level of smoking amongst 16-24 year olds from 26.5% in 2006 to 22.9% in 2012

These targets for the adult population are monitored via the Scottish Household Survey. ¹⁷ There are additional targets for children aged 13 and 15 which aim to reduce smoking levels by around a third by 2014. These will be measured via the Scottish Schools Adolescent Lifestyle and Substance Use Survey.

In 2007 the Scottish Government published Better Health, Better Care, ¹⁸ its action plan for improving health and health care in Scotland. This set out how NHS Scotland's HEAT¹⁹ performance management system (based around a series of targets against which the performance of its individual Boards are measured) would feed into the Government's overarching objectives. For example, one of the current HEAT targets is:

Through smoking cessation services, support 8% of each NHS Board's smoking population in successfully quitting (at one month post quit) over the period 2008/09 - 2010/11

The Scottish Government is committed to reducing inequalities in health, as outlined in Better Health, Better Care, ¹⁸ and the 2008 publications: Better coronary heart disease and stroke care: a consultation document²⁰ and Equally Well, ²¹ the report of the Ministerial Taskforce on Health Inequalities. The previous Scottish Health Surveys, as well as the Scottish Household Survey series, show that smoking levels are highest among those with the lowest incomes and among those living in the most deprived areas. Smoking-related mortality also shows a clear deprivation gradient: in 2004/5 32% of deaths in the most deprived 20% of areas in Scotland were attributed to smoking compared with 15% in the least deprived 20% of areas. ⁶ In light of this, a number of the anti-smoking actions and recommendations set out in these three policy documents, and those preceding them, ^22,23 relate specifically to disadvantaged groups, such as the development of pilot smoking cessation initiatives aimed at young people living in disadvantaged areas, the increasing move to screen people for cardiovascular disease risk factors such as smoking, and early years initiatives aimed at reducing smoking rates among pregnant women.

Although the population level targets for adults (16 and over) are monitored through the Scottish Household Survey, the Scottish Health Survey has an important role to play in terms of its ability to report on smoking as one of a range of cardiovascular disease risk factors. Scottish Health Survey data are also linked to the system of Scottish Morbidity Records, with participants' consent, providing a unique contribution to the study of smoking behaviour and later health outcomes.

This chapter will concentrate on smoking in adults aged 16 and over looking at the prevalence of current smoking, levels of smoking among current smokers, socio-demographic differences in smoking behaviour, second-hand smoke exposure and the receipt of medical advice to stop smoking. Trend data will also be presented by comparing data across the 1995, 1998, 2003 and 2008 Scottish Health Surveys.

4.2 METHODS

4.2.1 Smoking questions in the 2008 Scottish Health Survey

The questions on cigarette smoking were the same as those included in the previous three surveys, with some omissions.

As in previous years information about cigarette smoking was collected from adults aged 16 and 17 by means of a self-completion questionnaire which offered them the privacy to answer without disclosing their smoking behaviour in front of other household members. For adults aged 20 and over it was collected as part of the main interview. Those aged 18 and 19, at the interviewers' discretion, could answer the questions in the self-completion booklet or the main interview.

For young adults, the smoking questions in the self-completion questionnaire focus upon:

• 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.

The self-completion and main interview questions are mostly similar. However the main interview also asked about past smoking behaviour, desire to give-up smoking and medical advice to stop smoking.

In order to make way for other topics in the study a number of items were dropped in 2008. These include questions about smoking in pregnancy (dropped due to very low numbers of pregnant women within the sample), cigarette tar levels, health conditions that may have precipitated a quit attempt, and detailed questions about cigar and pipe smoking. Children aged 11-15 were asked about their smoking behaviour in 1998 and 2003. However, these questions were discontinued in 2008 and the Scottish Schools Adolescent Lifestyle and Substance Use Survey ( SALSUS) is now the only source of data on smoking among this age group in Scotland.

4.2.2 Cotinine

All those aged 16 years and over who were visited by the nurse were asked to provide a saliva sample in order to measure cotinine levels. Cotinine is an objective indicator of cigarette smoking - a level of 15ng/ml indicates that the person is a regular and/or recent smoker, but will not misclassify non-smokers exposed to second-hand smoke.

The measurement of cotinine levels in the Scottish Health Survey series provides an objective cross-check on self-reports of smoking behaviour, which are known to under-estimate prevalence. Inaccuracies in reporting arise in part from difficulties participants may experience in providing quantitative summaries of variable behaviour patterns, but in some cases arise from a desire to conceal the truth from other people, such as household members who may be present during the interview. This study is the only data source in Scotland which can provide a validated measure of self-reported smoking in this way. These data will be reported in 2010 when the results of the data collected in 2008 and 2009 via the nurse visit are published.

4.2.3 Definitions

The tables reported in this chapter use the following classifications of smoking status:

• Current smoking status: current smokers, ex-regular smokers, ex-occasional smokers, never smoked at all.
• Number of cigarettes smoked by current smokers: Light (under 10 cigarettes per day), moderate (10 to fewer than 20 cigarettes per day), heavy (20 or more cigarettes per day), and number of cigarettes not known. The mean number of cigarettes smoked is also used.

Unlike previous years, there is no longer a distinction between 'current cigarette smokers' and 'current smokers' (the latter including those who smoke pipes or cigars in addition to those who only smoke cigarettes). As a result of the question changes, the tables now report only on current cigarette smokers.

4.3 TRENDS IN SMOKING PREVALENCE AND EXPOSURE TO SECOND-HAND SMOKE

4.3.1 Trends in smoking prevalence, by age and sex

Table 4.1 shows the percentages for current cigarette smokers, ex-regular cigarette smokers and those who were never regular cigarette smokers/never smoked at all, for the four years the SHeS has been conducted. As there have been changes to the sample over the years the total figures are based only on those aged 16 to 64. The steady declines in current smoking prevalence seen in previous survey years have continued through to 2008. Falling from 36% in 1998 to 32% in 2003, the figure has fallen further to 29% among men aged 16 to 64. The consistent declines among women aged 16 to 64 from 36% in 1995, to 33% in 1998 and to 31% in 2003 have continued with a further drop to the most recent level of 28% in 2008. The levels of ex-regular cigarette smoking have stayed largely the same among men (18% in 1995 to 19% in 2008) while this has increased slightly among women, from 16% in 1995 to 19% in 2008. The proportion of people who were never regular cigarette smokers, or who never smoked at all, increased steadily over the survey years from 45% in 1995 to 51% in 2008 among men and from 49% in 1995 to 53% in 2008 among women. The overall mean number of cigarettes smoked per day by current smokers steadily decreased from 18.1 cigarettes in 1995 to 15.6 cigarettes in 2008 among men and from 15.4 cigarettes in 1995 to 13.6 cigarettes in 2008 among women. Trends were generally the same for all age groups. Table 4.1

4.3.2 Trends in non-smokers' exposure to second-hand-smoke by age and sex

Non-smokers' exposure to second-hand smoke has declined markedly in the decade since 1998, as shown in Table 4.2. Exposure to second-hand smoke in the participants' own home has decreased from 18% to 10% in both men and women, with decreases consistently applying across all age-groups. Second-hand smoke exposure in other people's homes has decreased from 21% in 1998 to 12% in 2008 among men and from 25% in 1998 to 13% in 2008 among women, generally applying across age-groups. In addition to these declines in domestic settings, with the introduction of the ban on smoking in public places, exposure in pubs and on public transport has been virtually eliminated. In 2008 just 1% of men and less than 1% of women reported being exposed to smoke in these places, the comparable figures for pubs in 1998 were 44% and 30%, respectively, while for public transport the 1998 figures were 7% and 8%. Dramatic falls have also been seen in proportions exposed in other public places and at work. For example, around a quarter of men in 1998 were exposed to smoke at work or in public places (other than on transport or in pubs) compared with just 6% in 2008. Women's exposure to smoke at work declined in the same period from 14% to 2%, while their exposure in public places (other than on transport or in pubs) fell from 28% to 6%. Proportions not exposed to other people's smoke have doubled from 33% in 1998 to 73% in 2008 among men and from 39% in 1998 to 75% in 2008 among women. Table 4.2

4.4 SMOKING PREVALENCE IN 2008

4.4.1 Smoking prevalence, by age and sex

Current cigarette smoking prevalence for all adults aged 16 and over in 2008 was 26%; 27% for men and 25% for women, as shown in Table 4.3. Figures were significantly different between men and women and across age groups; the most notable difference being that women aged 55 and over were more likely than men of this age to have never smoked at all, and the gender gap for occasional ex-smoking in those over 65 was also reasonably large. Current smoking rose from 28% in men aged 16-24 to a peak of 36% in those aged 25-34 before falling consistently with increasing age to 10% in those aged 75 and over. Current smoking prevalence was around 29% in women aged 16 to 54, which decreased with age to 11% in those aged 75 and over. Proportions who had never smoked cigarettes at all were 45% for men and 48% for women and tended to be higher for the younger ages and lower for the older ages. Ex-regular smoking levels were 24% for men and 22% for women, which generally increased with age. The prevalence of heavy smoking (20 or more cigarettes per day) for the entire population (including non-smokers) was 9% for men and 7% for women but higher among those aged 35 to 64. Again, differences were significant by sex and across age groups.

The overall cigarette smoking level of 26% for all adults (27% for men and 25% for women) is still some way short of the target level of 22% by 2010. ¹⁵ The Scottish Household Survey has a longer time series of data for the adult population than the Scottish Health Survey and is therefore used to measure progress towards this target. The 2008 Scottish Household Survey figures are similar to those from the Health Survey: 25.2% overall, 24.9% for men and 25.4% for women (these very small differences between the two surveys are not statistically significant). Table 4.3, Figure 4A

Figure 4A Current cigarette smoking by age and sex

4.4.2 Number of cigarettes smoked by age and sex

Table 4.4 presents the mean number of cigarettes smoked per day among smokers and self-reported daily cigarette consumption. The mean number of cigarettes smoked per day among smokers was 15.7 for men and 13.7 for women, but the figures were higher for those aged 35 to 74. Among smokers, the heavy smoking prevalence was 34% for men and 28% for women; figures were higher for those aged 35 to 74. 24% of male and 30% of female smokers were light consumers with figures generally higher for the younger groups for both sexes, and lower for those aged 35 and older among men, and those aged 35 to 64 among women. These differences between men and women and across age groups were significant. Table 4.4

4.4.3 Smoking prevalence by socio-demographic factors

Tables 4.5 to 4.7 present the prevalence of smoking by socio-economic classification ( NS-SEC of the household reference person), equivalised household income and the Scottish Index of Multiple Deprivation (descriptions of each of these measures are available in the Glossary at the end of this volume). To ensure that the comparisons presented in this section are not confounded by the different age profiles of the sub-groups, the data have been age-standardised (for a description of age-standardisation please refer to the Glossary). On the whole the differences between observed and age-standardised percentages are small. Therefore, the percentages and means presented in the text below are standardised. The tables report both the observed and the age-standardised figures.

Socio-economic classification ( NS-SEC)

Table 4.5 and Figure 4B present current smoking levels by the occupational status of the household reference person in terms of NS-SEC. Current smoking variations by NS-SEC were statistically significant in both men and women. Levels were highest among men and women in semi-routine and routine households and lowest among those in managerial and professional households, with the prevalence of smoking in the former group being double that in the latter. The age-standardised prevalence of current smoking in these two groups was 36% and 17% among men and 38% and 17% among women, respectively. The pattern of the association between current smoking status and socio-economic group was different for men and women. In men, the smoking prevalence among managerial and professional households was significantly lower (17%) than in the other groups (26%-36%), whereas for women the smoking prevalence among those in semi-routine and routine households (38%) was almost double that in the other groups (17%-22%). The mean number of cigarettes smoked per smoker tended to follow a similar pattern to that for current smokers, with lower consumption levels found in managerial and professional households (12.8 cigarettes for men and 11.8 cigarettes for women) than in semi-routine and routine households (16.7 and 14.8 cigarettes, respectively), although the highest number of 17.2 cigarettes was seen for men in lower supervisory and technical households.

Patterns of never smoking mirrored those for current smoking with figures dropping from 53% among men and 56% among women in managerial and professional groups to 38% and 35%, respectively, in semi-routine and routine households. Once again, the patterning of never smoking was different in men and women: in men the percentage among managerial and professional households (53%) was somewhat distinct from that in other groups (38%-45%), whereas for women the most pronounced distinction in prevalence was between those in semi-routine and routine households (35%) and other groups (48%-56%). There were no clear patterns in relation to the percentage who used to smoke cigarettes either occasionally or regularly for either sex, though men in lower supervisory and technical and semi-routine and manual households were the least likely to have been occasional smokers. Table 4.5, Figure 4B

Figure 4B Current cigarette smoking (age-standardised), by NS-SEC of household reference person and sex

Equivalised household income

Table 4.6 shows that the age-standardised prevalence of current smoking increased steadily with decreasing equivalised household income, with the prevalence more than doubling between the highest and lowest quintiles for both men and women. 18% of men in the highest income quintile households currently smoked compared with 44% of those in the lowest income quintile households. The corresponding figures for women were 16% and 39%. Among men, the mean number of cigarettes smoked per day by current smokers was highest among those in the 4 ^th and 5 ^th (lowest) quintiles (16.8 and 16.6, respectively) while figures were lower but relatively constant across the remaining three groups (ranging between 14.9 and 15.1). The equivalent figures for women showed a more linear increase as income declined, from 11.9 cigarettes among women smokers in the highest income households to 15.4 cigarettes for those in the lowest income households.

The proportions who had never smoked fell steadily from 53% of men and 56% of women living in households in the highest income quintile to 34% and 37% in the lowest quintile respectively. Men in the two lowest income quintiles were also the least likely to have been occasional smokers. There were no consistent patterns for women in relation to occasional smoking, or in relation to ex-regular smoking for either sex. Table 4.6, Figure 4C

Figure 4C Current cigarette smoking (age-standardised), by equivalised household income quintile and sex

Scottish Index of Multiple Deprivation ( SIMD)

Two measures of SIMD are being used throughout this report. The first, which uses quintiles, enables comparisons to be drawn between the most and least deprived 20% of areas and the intermediate three quintiles, and helps to assess the extent of any inequalities in behaviours or outcomes. The second contrasts the most deprived 15% of areas with the rest of Scotland (described in the tables as the "85% least deprived areas"). The most deprived 15% of areas have been identified as of particular concern by Scottish Government and are the subject of a number of policy initiatives. The Scottish Health Survey is designed to provide a large enough sample in the SIMD 15% areas after four years to enable robust detailed analysis at this level.

Significant gradients were seen for both men and women in current smoking prevalence by area deprivation as measured by SIMD quintile, as shown in Table 4.7. Among men, the age-standardised prevalence of current smoking rose from 15% in the least deprived areas to 39% in the most deprived areas; the corresponding figures for women increased from 14% to 39%. Mean cigarettes smoked per smoker generally increased with increasing area deprivation, from 11.0 cigarettes in men and 12.1 cigarettes in women living in the least deprived areas to 17.6 cigarettes in men and 15.4 cigarettes in women in the most deprived areas. Conversely, the proportions who had never smoked steadily decreased with increasing area deprivation from 57% of men and 60% of women living in the least deprived areas to 33% of men and 36% of women living in the most deprived areas. Ex-occasional smoking decreased with increasing area deprivation among women but not men and there were no clear patterns in the levels of ex-regular smoking by area deprivation.

The differences in the prevalence of current smoking between those in the most deprived 15% of areas and the rest of Scotland were large for both men and women (42% versus 24% for men and 41% versus 22% for women). However, the analysis by SIMD quintile reveals the full extent of the sharp inequalities in smoking behaviour between those in the most and least deprived areas of Scotland. Table 4.7, Figure 4D

Figure 4D Current cigarette smoking (age-standardised), by Scottish Index of Multiple Deprivation quintile and sex

4.5 MEDICAL ADVICE TO GIVE UP SMOKING

The proportions of smokers who have ever received medical advice to give up smoking by age are shown in Table 4.8. 34% of men and women had received advice at some point, 14% had done so in the last year and the remainder longer ago than that. The percentages varied significantly by age, and although there were no clear general patterns, overall proportions were highest among men aged 55-64 (40%) and 65-74 (41%), and were lowest among women aged 75 and over (21%). However, men and women aged 55 and above were more likely to have received their advice to quit more than a year ago than in the previous year, whereas the balance tended to be more evenly split for the age groups below this. Table 4.8

4.6 EX-SMOKERS

Of those who had stopped smoking, 4% of men and 7% of women had done so in the six months prior to survey interview (Table 4.9). The figures extended to 9% of men and 12% of women when considering those who had stopped within 2 years. Around 41% of men and 40% of women had stopped regular smoking 20 or more years ago. People aged under 55 were more likely to have given up smoking in the period of time since the ban in 2006 than those aged 55 and over; the majority of this older age group had stopped smoking 20 or more years ago. Table 4.9

4.7 FACTORS ASSOCIATED WITH SMOKING

Logistic regression was used to examine the association between current smoking prevalence and socio-demographic, behavioural, and anthropometric factors. The dependent variable was current smoking. By simultaneously controlling for a number of factors, the independent effect each factor has on the variable of interest can be established. Logistic regression compares the odds of a reference category (shown in the table with a value of 1) with that of the other categories. Separate logistic regressions were run for men and women. The following independent variables were found to be significant for both sexes: age group, NS-SEC, equivalised household income, SIMD, marital status, body mass index group, fruit and vegetable consumption and alcohol intake. Parental NS-SEC and physical activity level were also investigated but found not to be associated with current smoking.

The odds ratios for current smoking are presented in Table 4.10. An odds ratio of less than one means that the odds of a given group currently smoking are lower than that of the reference category. The opposite is true when the odds ratio is greater than one. Independent variables with a 'p' value of 0.05 or less are significant predictors of the dependent variable at the 95% confidence level.

Once all other factors were controlled for, men aged 25 to 44 and women aged 25 to 54 had significantly higher odds of currently smoking than those aged 16-24, while those aged 75 and over had significantly lower odds. For example, the odds ratios for men and women aged 75 and over were 0.31 and 0.47, respectively.

Men in small employer and own account worker households, lower supervisory and technical and in semi-routine and routine households had significantly higher odds of currently smoking than those in managerial and professional households (odds ratios of 1.62, 1.69 and 1.74 respectively). Women in semi-routine and routine households also had significantly higher odds of currently smoking than those in managerial and professional households (odds ratio of 1.71).

Men living in households in the 4 ^th and 5 ^th (lowest) income quintile had significantly higher odds of currently smoking than those living in households in the highest income quintile (odds ratios of 1.58 and 1.92, respectively). The odds of women smoking increased significantly in each income quintile relative to those in the highest group; from 1.44 in the 2 ^nd to 1.97 in the 5 ^th lowest quintile.

There was a steady increase in the odds of current smoking with increasing area deprivation as measured by SIMD, with significant differences in both men and women between the 2 ^nd and 1 ^st (most deprived) quintile and the least deprived quintile; additionally women living in the 3 ^rd quintile also had significantly increased odds. The odds ratios for those in the 1 ^st (most deprived) quintile were 1.84 for men and 2.43 for women.

Men and women who were living as married, separated or divorced had significantly higher odds of currently smoking than their married counterparts. Additionally, the relative odds for widowed men and single women were also higher.

Compared with people of normal weight ( BMI <25 kg/m ²), overweight ( BMI 25 to <30 kg/m ²) and obese people ( BMI 30 kg/m ² and over) were at decreased odds of current smoking (odds ratios of 0.70 and 0.74 among overweight men and women, respectively; 0.42 and 0.66 among those who were obese, respectively).

Compared with those eating 5 or more portions of fruit and vegetables per day, men and women eating fewer than 5 portions per day had significantly increased odds of currently smoking (odds ratios of 1.65 and 1.79, respectively). The effect was even more pronounced for those eating none (odds ratios of 3.12 and 3.46, respectively).

Finally, men and women exceeding the recommended weekly alcohol consumption guidelines for their sex had significantly raised odds of currently smoking compared with those consuming less than the recommended amounts (odds ratios of 1.67 for men and 2.23 for women).

This form of analysis cannot reveal the direction of causation for these associations, however it underlines two things. Firstly, it confirms the extent to which smoking behaviour is heavily socially patterned, with NS-SEC, income and area deprivation all demonstrating significant independent associations. Secondly, it highlights the extent to which risk factors for poor health outcomes such as smoking, low fruit and vegetable consumption and excess alcohol consumption all co-exist. Table 4.10