Rachel Turner
|
Summary
|
As stated in a policy document issued by the Scottish Office, smoking ‘is the single most important contributor to ill-health in Scotland.’1 The Scottish Office has set targets to reduce the prevalence of smoking in Scotland from the 1986 levels of 30% for 12-24 year olds and 40% for 25-65 year olds to 21% and 32% respectively by the year 2000.2 As well as action taken to reduce smoking prevalence at the UK level, for example through restrictions on advertisements for tobacco products, additional measures are taken in Scotland through health education programmes by the Health Education Board for Scotland and by individual Health Boards, voluntary organisations, employers and local authorities.3
The Scottish Health Survey has collected smoking data using identical questions to those used in the General Household Survey (GHS) and in the series of Health Surveys for England.4 The questions focused on cigarette smoking, whether ready-made or roll-ups, although questions about the prevalence of pipe and cigar smoking were also included.
The data on smoking for informants aged 18-64 was collected by means of a face-to-face interview. Those aged 16 and 17 were asked by the interviewer to complete a self-completion questionnaire. Of course, in either case the presence of other household members in the same room may have had an effect on an informant’s responses.
Those aged 16 and 17 were only asked a sub-set of the smoking questions: whether they currently smoked and if so how many, at what age they had started smoking, whether they had tried to give up, and if they were exposed to smoke in particular places. Informants aged 18-64 were asked these questions plus additional ones including (if relevant) how long ago they had stopped smoking, the reasons why they stopped, if they smoked during pregnancy, and if they had ever received medical advice to stop. Serum cotinine was analysed for all informants who gave a blood sample during the nurse visit.
This chapter is divided into nine substantive sections. Sections 4.2 and 4.3 analyse smoking prevalence and the number of cigarettes smoked by age and sex. Section 4.4 looks at past cigarette smoking (for informants aged 18-64). Section 4.5 analyses serum cotinine levels in the blood samples provided by informants. Section 4.6 looks at passive smoking. Sections 4.7 and 4.8 analyse smoking prevalence by social class and region respectively. Section 4.9 compares smoking prevalence in Scotland with that in England, using data from the 1995 Health Survey for England.5 Finally, Section 4.10 compares Health Survey data with the earlier Scottish Heart Health Study.
4.2 SELF-REPORTED SMOKING PREVALENCE BY SEX AND AGE
Throughout this chapter a distinction is made between ‘current smokers’ and ‘current cigarette smokers’: whereas the latter refers to those who said they smoked cigarettes at all at the time of the interview, the former also includes those who said they smoked cigars or a pipe. Some of these informants smoked very little - for example, 2% smoked on average less than one cigarette a day.
Just over one in three informants were current cigarette smokers: 34% of men and 36% of women (a difference which was not statistically significant). Women were slightly more likely than men to say they never regularly smoked cigarettes: 49% women and 45% men. Men, on the other hand, were more likely than women to say they had been regular smokers in the past: 21% men and 16% women. Thus men were more likely than women to have ever been regular smokers (i.e., current regular smokers plus ex-regular smokers): 55% men and 51% women. Table 4.1
Six percent of men reported that they smoked a pipe or cigar but not cigarettes; the proportion of women who fell into this category was negligible. Thus, the inclusion of pipe and cigar smoking reversed the difference between men and women, with men significantly more likely than women to smoke: 39% of men compared with 36% of women were thus classified as current smokers. (Section 4.5 presents evidence from blood analyses which suggests that there was some under-reporting of current cigarette smoking, and that men were more likely than women to under-report.)
For five age groups, Figure 4A shows the proportion of men and women who reported smoking cigarettes and, for men only, the proportion smoking at all (cigarettes, pipe, cigar). The proportion of current cigarette smokers varied by age but no clear pattern emerged. For both men and women the age group most likely to report that they currently smoked cigarettes was 25-34 years. Figure 4A
Figure 4A: Proportion of men and women currently smoking, by age
When pipes and cigars are included, the proportion of men reporting that they smoked at all (cigarettes, pipe, cigar) tended to increase with age. As very few men in the youngest age group said they smoked a pipe or cigar but not cigarettes, the proportion of men under age 35 classed as smokers was not greatly affected by the inclusion of pipe and cigar smoking. However, starting with the 35-44 age group, the proportion of men smoking increased by about seven to nine percentage points when pipes and cigars were included. Thus, the increase in smoking with age was due to the higher proportion of men in the older age groups who smoked pipes or cigars. With the inclusion of pipe and cigar smoking, men were more likely than women to report smoking in all age groups and, as age increased, the difference between the proportions of men and women smokers also increased.
The Scottish Office target for smoking prevalence for 25-65 year olds is 32% by the year 2000; the current level (for 25-64 year olds) measured by the Health Survey is somewhat higher at 35%.
4.3 NUMBER OF CIGARETTES SMOKED BY AGE AND SEX
In order to derive estimates of the number of cigarettes smoked on an average day, current cigarette smokers were asked how many cigarettes they usually smoked a day on weekdays and how many per day they usually smoked at weekends. The weekday figure was then multiplied by five and the weekend figure by two, and total weekly consumption was estimated by adding these two figures together. This sum was then divided by seven in order to estimate average daily consumption.
4.3.2 Number of cigarettes smoked
Among current cigarette smokers, 41% reported smoking 20 or more a day (18% smoked 20, 14% between 20 and 30, and 9% smoked 30 or more). By contrast, 8% reported smoking less than five a day, and 2% smoked less than one a day.
Sex differences
There was a significant difference between the proportions of men and women smoking 20 or more cigarettes a day. Forty-five percent of male smokers said they smoked 20 or more cigarettes a day (15% smoked 20, 17% between 20 and 30, and 14% said 30 or more) compared with 37% of female smokers (20% smoked 20, 12% between 20 and 30, and 5% said 30 or more). Based on the total sample of 16-64 year olds, rather than on current cigarette smokers only, 15% of men and 13% of women smoked 20 or more cigarettes a day. Table 4.2
Men smoked, on average, significantly more cigarettes per day than women: 18.1 compared with 15.4. The median number of cigarettes smoked was also higher for men than women: 17 and 15 respectively.
By giving current non-smokers a value of zero cigarettes per day, it becomes possible to calculate average consumption for the whole sample: on this base, the overall mean for men was 6.1, while for women it was 5.5
Age differences
The average number of cigarettes smoked per day is shown in Figure 4B by age for men and women smokers. As can be seen, the average increased with age until the 45-54 age group, after which it fell slightly for women and stayed the same for men. Figure 4B
Figure 4B: Average number of cigarettes smoked per day, by age
4.4.1 Whether ever smoked cigarettes
Overall, 43% of men and 49% of women said that they had never regularly smoked cigarettes (‘regularly’ is at least one per day). When current pipe or cigar smokers who had never regularly smoked cigarettes are included, the total increases to 45% of men but stays the same for women. Table 4.1
A further 18% of men and 16% of women had been regular cigarette smokers in the past, but had stopped by the time of the Health Survey. If current cigar or pipe smokers who had previously been regular cigarette smokers are added in, the total of past regular cigarette smokers increases to 21% for men, but remains the same for women at 16%.
It is not surprising to find that the proportion of ex-regular cigarette smokers increased with age for both men and women. Thus, among men aged 45-64, almost seven out of ten (68%) had regularly smoked cigarettes at some time (i.e., including current and ex-regular smokers). But of the men aged 16-44, only 48% had ever been regular cigarette smokers. Of women currently aged 45-64, 57% had smoked cigarettes regularly at some time, compared with 48% of those aged 16-44. Figure 4C
Figure 4C: Past and present cigarette smoking, by current age
4.4.2 How long ago stopped smoking
As the self-completion questionnaire answered by informants aged 16 and 17 did not cover the remaining topics in Section 4.4, the following analyses are restricted to those aged 18-64.
Ex-regular cigarette smokers were asked how long ago they had stopped smoking. Women were more likely than men to have stopped within the past five years (35% compared with 30%). This difference is explained by the higher proportion of ex-smokers among older men who gave up long ago, rather than by women being inclined at the present time to give up smoking at a faster rate than men. There was a significant difference between the percentage of men and women who had stopped smoking 10 or more years ago: 53% of male ex-regular smokers had given up 10 or more years ago, compared with 47% of women. Table 4.3
A quite high proportion of men and women ex-smokers had given up smoking in the past twelve months: 11% of men and 9% of women. Although the analysis that can be done with this group of recent ex-smokers is limited because of their small numbers in the sample, it appears that women aged 18-24 were the group most likely to have given up smoking in the twelve month period before the interview. As Table 4.4 shows, out of all women who had given up smoking in the past twelve months, the largest group were the 18-24 year olds; out of all men who had given up smoking during this period, the largest group were the 25-34 year olds. Table 4.4
4.4.3 Medical advice to stop smoking
Current and ex-cigarette smokers (aged 18-64) were asked if they had ever been advised to stop smoking by a doctor or nurse. The likelihood of being advised to stop smoking increased with age for men and, to a lesser extent, for women after age 45. However, the group most likely to have been given such advice within the past year was women aged 18-24; as described in the previous section, it was precisely this group who were also the most likely to have stopped smoking in the year before the interview. Table 4.5
Among men, 36% of current cigarette smokers said they had received medical advice to stop smoking, as did 21% of ex-regular cigarette smokers, and 5% of those who had smoked cigarettes occasionally. The figures for women were similar. However, among current cigarette smokers, women were more likely than men to have been advised in the past year to stop smoking (21% compared with 16%).
Among current smokers, the likelihood of being advised to stop was directly related to the number of cigarettes smoked per day. For example, the proportion advised to stop smoking was 19% among men who currently smoked less than 10 a day, while it was 34% of men smoking between 10 and 20 a day and 44% of men smoking 20 or more a day. The pattern among current women smokers was similar, with the proportions being advised to stop in each of these three categories slightly higher than for men Table 4.6
4.4.4 Whether tried to give up smoking due to a health condition
Twenty-seven percent of those who were current or ex-cigarette smokers (including regular and occasional ex-smokers) said they had tried at some point to give up smoking because of a health condition. Women were more likely than men to have tried to give up smoking for health reasons (34% compared with 21%).
Those who had tried to give up smoking for health reasons were asked to specify the particular condition concerned. Half (50%) of all women who said they tried to give up smoking as a result of a health condition mentioned pregnancy. It is worth noting that, as pregnancy might not be regarded by all women as a ‘health condition’, this may be an underestimate of the proportion of women who try to stop smoking because of pregnancy.
Table 4.7 shows the health conditions mentioned by informants (other than pregnancy). Informants could mention up to eight health conditions (from a list of twelve) as possible reasons for their attempts to give up smoking. In Table 4.7, three of the categories of conditions subsume a number of those from the initial list: ‘cardiovascular condition’ includes both heart problem and high blood pressure; ‘respiratory problem’ includes bronchitis, cough, shortness of breath and other respiratory problems (including asthma); and ‘other health reason' includes ulcer and other gastro-intestinal problems, diabetes and any other health conditions not included on the list. The figures shown in the table refer to those mentioning any of the conditions within the relevant category. Table 4.7
Respiratory problems were the category of conditions mentioned most frequently. Over three in five of those who had tried to stop smoking mentioned them (62% of men and 65% of women). Among men, the next largest category (though much smaller) was cardiovascular conditions, which increased with age and peaked at 36% of those aged 55-64. A similar, but less marked, increase by age was found among women who mentioned cardiovascular conditions.
Women who were either current cigarette smokers, or who had given up smoking in the past year and who were currently pregnant or had been pregnant in the past twelve months, were asked if they had smoked at all during their pregnancy. Ninety-eight women fell into this category. The small group answering these questions does not allow detailed analysis. (The figures quoted below are unweighted.)
Out of this group, twelve women were currently pregnant. Ten of these twelve had not smoked at all since they had known they were pregnant and two had smoked some of the time. Of these twelve, eight said they either stopped smoking or only smoked some of the time because of the pregnancy, and the other four had done so for some other reason.
Eighty-six women who fell into the group of current smokers or recent ex-smokers had been pregnant in the previous twelve months. Out of these 86, 53 of them reported that they had smoked all the time during their pregnancy, 16 had smoked some of the time and 17 had not smoked at all during their pregnancy. Of those who smoked some of the time or not at all during their pregnancy, 25 had done so because of the pregnancy and 7 for some other reason (one person did not answer the question).
4.4.6 Age at which started regular cigarette smoking
Informants aged 18-64 who were current or ex-regular cigarette smokers (regular being defined as at least one cigarette a day) were asked at what age they had first started smoking.
Over four in five cigarette smokers (81% of male, and 82% of female, smokers) said they started smoking regularly between the ages of 14 and 24. About half of smokers had started between the ages of 14 and 17 (53% of male smokers and 51% of female smokers). Table 4.8
In general, women claimed to have started smoking somewhat later than men. Significantly more women than men had started smoking at age 20 or later: 16% of men compared with 21% of women. And the proportion of informants who started smoking before they were 14 years old was 15% for men and 11% for women.
Of course, this analysis is inevitably somewhat confounded by the fact that some of the younger non-smoking informants are likely to take up cigarette smoking later in their lives. However, as Table 4.8 shows, women aged 45-64 tended to start smoking cigarettes at a later age than women aged under 45: 35% of the former started smoking at or over the age of 20, but only 13% of the latter had done so. It seems unlikely that there will be enough new smokers in future in the current 25-44 age group to bring the proportion of women starting after age 20 up to 35%. Of women now aged 45-64, 6% said they began smoking before the age of 14, which is much lower than the 15% of women aged 18-44 who started by age 14. It appears, therefore, that women are taking up smoking at an earlier age nowadays than they did in earlier decades.
The pattern for men was somewhat different. Although more men aged 45-64 claimed that they had started smoking at or over the age of 20 compared with men aged 25-44 (20% and 15% respectively), it is not clear that in the future the current 25-44 age group could not bring the proportion of those starting at 20 or over to a similar percentage. Furthermore, the same proportion of men aged 45-64 had started smoking before age 14 as current 18-44 year olds (15% for both age groups). So, unlike the evidence for women, it does not appear that men are starting to smoke at an earlier age nowadays.
For the younger age group (18-44), the pattern for men and women was similar: 15% of both sexes had started smoking before the age of 14. By contrast, among those aged 45-64, men were much more likely than women to begin smoking before the age of 14 (15% of men compared with 6% of women).
Obtaining accurate information on smoking behaviour can be a difficult task to undertake in a survey which relies on self-reports of what may be considered by some people to be a ‘sensitive’ activity. There are several reasons why this may be so. Firstly, when interviews take place in informants' own homes, they may not wish to reveal whether, or how much, they smoke in front of other household or family members. Similar considerations might also apply in other settings - for example, in clinical settings, informants may wish to conceal their actual behaviour from their own doctor because of medical advice they may have been previously given or because of worries in getting insurance cover, and so on. Secondly, even if there is no intention to conceal the truth, informants may still experience difficulties in estimating average levels of consumption for a behaviour that most likely varies both day to day and over longer periods of time.
Therefore, in order to improve estimates on levels of smoking, a number of biological markers have been developed as potential indicators of active and/or passive smoking (particularly among children). Some of these markers include carbon monoxide, thiocyante and cotinine (which is a metabolite of nicotine).6 While no marker is ideal for this purpose, cotinine has generally been considered the most useful by experts.7
The continine analysis in the Health Survey was based on a blood sample taken during the nurse visit. The Health Survey used gas chromatography, which allows very low concentrations of serum to be detected. (Further details about all blood analytes may be found in the Technical Report.)
Previous research suggests that serum cotinine levels above (approximately) 20 nanograms per millilitre (ng/ml) provide good evidence that the person is a smoker; on the other hand, levels of (approximately) 8 ng/ml or less indicate a non-smoker. Analysis of serum cotinine levels therefore has potential to be used as an external validation of self-reported smoking behaviour.
As nicotine replacement products (including patches, chewing gums and nasal sprays) are known to affect levels of serum cotinine, the small number of people who had used such products within seven days of the nurse visit have been excluded from the analysis described below.8
When those on nicotine replacement products are removed, cotinine assays were recorded for 5,763 respondents (which is 73% of the total unweighted sample). The youngest age group (16-24) was slightly under-represented in the assay sample (17% among men and 15% among women) compared with the total sample (19% and 17% respectively). Table 4.9
Serum cotinine levels ranged from under 0.1 ng/ml to over 700 ng/ml. There was a very large cluster of low values, which tended to be well below 8 ng/ml and was mainly found among self-reported non-smokers. There was also a smaller cluster of values well above 20 ng/ml, which were largely found among self-reported smokers.
Overall, 61% of informants had serum cotinine levels below 8 ng/ml (60% of men and 62% of women). Forty-eight percent of all measured levels were very low, that is below 1.5 ng/ml. Only 1% of serum cotinine levels fell between 8 and 20 ng/ml (that is, between the upper limit of the non-smoker threshold and the lower limit of the smoker threshold).
Above 20 ng/ml, informants were divided into four bands of approximately equal size (although the lowest of these was slightly smaller than the other three): 20 to less than 150 ng/ml (18% of all 20 ng/ml or over), 150 to less than 250 (26%), 250 to less than 350 (27%), and 350 or over (29%).
4.5.3 Serum cotinine levels by self-reported smoking status
Current smokers
Among those informants who said they were current cigarette smokers, 3% had serum cotinine levels under 8 ng/ml (that is, below the threshold normally identified for smokers). Most (81%) of this group were estimated to smoke fewer than five cigarettes a day on average, and almost half (48%) averaged less than one a day.
Thus, the low serum cotinine levels for this group could be explained by this apparently low average consumption among them. Another possible reason could be that, given that serum cotinine has a short half-life (of 16-20 hours), the smoking behaviour of the these informants was untypical in the day or two before the nurse visit.
Not a current cigarette smoker
Among those who said they did not currently smoke cigarettes, 91% had serum cotinine levels below 8 ng/ml, 2% had levels of 8 ng/ml to less than 20 ng/ml, and 7% had levels of 20 ng/ml or more. Among the latter, 40% currently smoked a pipe or cigars, which may have accounted for their having a serum cotinine value of 20 ng/ml or more. (Almost three quarters of the pipe or cigar smokers had in the past smoked cigarettes regularly.)
Previous research suggests that informants with a serum cotinine value of 20 ng/ml or more who said that they did not smoke at all will most likely have been smokers. Using the terminology of the reports on the Health Survey for England, this group is referred to as 'misreporters'. There were 171 such cases (using the
weighted figures). Among these, 57% said they were ex-regular cigarette smokers, and a further 27% that they had previously smoked cigarettes occasionally (and 25% of these ex-smokers said they had given up within the past year). The misreporters were more likely to be men than women (60% and 40%); and 36% were under age 25 (compared with 18% of the total sample and 17% of current smokers).
If it is assumed that these 171 cases were in fact current smokers, a revised estimate can be made for the proportion of the population classified as smokers. The misreporters were classified by sex and age group. For each age group, the number of misreporters was expressed as a percentage of those classified as non-smokers. To project the assay sample results onto the total sample, the percentage derived from the assay sample was multiplied by the number of non-smokers in the full sample for that age group in order to estimate the number of additional smokers for each age group. This figure was then added to the number of self-reported smokers, and expressed as a percentage of the whole sample in the age group. Looking at all age groups, the adjusted estimate of smoking prevalence was 40% overall, 43% for men and 38% for women. Table 4.10
Because the misreporters were not distributed in proportion to the size of the different age/sex groups, the adjusted estimates for smoking prevalence by age and sex differ in some respects from the unadjusted estimates. The most likely to misreport were men aged 16-24: the proportion of 16-24 year old men who smoked increased by nine percentage points, but the increase was only one to three percentage points in other age groups. This results in a higher estimate of smoking prevalence among men aged 16-24 years than among those aged 25-34 years.
The proportion of women classed as smokers increased by two or three percentage points for all age groups except the youngest (16-24) where the increase was four percentage points.
Of course, as this adjustment deals only with those informants whose serum cotinine levels were within the range considered to be clearly indicative of a smoker (i.e., at least 20 ng/ml), it leaves out the possibility that some misreporters may have had levels below 20 ng/ml (as some self-reported smokers had levels below this). Regardless of which estimate is used, the unadjusted or the adjusted, it must in any case be borne in mind that some informants classed as current cigarette smokers had very low levels of consumption.
4.5.4 Serum cotinine levels by sex and age
When those on nicotine products are removed from the sample (but not misreporters), overall 36% of those from whom a cotinine assay sample was recorded had serum cotinine levels of 20 ng/ml or more (37% of men compared with 35% of women). These figures may be compared with the unadjusted figures for smoking prevalence (that is, those who reported smoking either cigarettes, pipes or cigars): 39% of men and 36% of women. When the adjusted figures are used, that is, when those who misreported (had cotinine levels of 20 ng/ml or more but reported not currently smoking) are included as current smokers, the proportion of smokers increased to 43% of men and 38% of women.
The mean level of serum cotinine was higher for men than women (104 compared with 93), which is consistent with self-reported data on mean number of cigarettes smoked which was also higher for men than women (18.1 per day compared with 15.4). Table 4.11
Figure 4D compares the proportion of men and women with serum cotinine levels of 20 ng/ml or more with the proportion of self-reported current cigarette smokers for each age group. For both sexes, the comparison shows that the proportions were very similar. Among men, the proportions with cotinine levels of 20 ng/ml or more decreased with age until the 35-44 age group, after which it increased and remained stable at 38% for the two oldest age groups (45-54 and 55-64). The larger difference between the proportion of men classed as current cigarette smokers and the proportion with cotinine levels of 20 ng/ml or more in the youngest age group (35% compared with 42%), can be accounted for by the higher proportion of misreporters in that age group compared with the others. Among women, cotinine levels of 20 ng/ml or more followed very closely reported cigarette smoking. Figure 4D
Figure 4D: Proportion with cotinine levels >20 ng/ml, proportion of self-reported current cigarette smokers, by age
Figure 4E shows, separately for men and women, mean serum cotinine level and mean number of cigarettes smoked per day (based on the whole sample), by age. Given that these both attempt to measure the volume of cigarette consumption, a relationship between these two measurements would be expected. As in the previous figure, the relationship between these two measures was very close. Among men, mean serum cotinine level increased between the first and second age groups, then decreased between the second and third before increasing again for those aged 45 or over. As the figure shows, the mean number of cigarettes smoked per day followed a similar pattern.
If the sample is divided into two age groups, mean serum cotinine levels were significantly different for men aged 16-44 (99.8) and those aged 45-64 (112.9).
Figure 4E: Mean serum cotinine level and mean number of cigarettes per day, by age
Among women, mean serum cotinine levels and mean number of cigarettes per day matched each other closely by age group. They both increased between the first and second age groups, decreased between the second and third, increased again slightly before decreasing again for those aged 55-64. However, there was little overall difference between mean serum cotinine levels of those aged 16-44 and those aged 45-64 (91.7 and 94.4 respectively).
4.5.5 Serum cotinine levels and current smoking behaviour
Table 4.12 looks at serum cotinine levels by self-reported smoking status. As misreporters (see 4.5.3 above) could distort the results within the smoking categories, they have been removed from this table.
For current smokers of both sexes, serum cotinine levels markedly increased along with the number of cigarettes smoked. For men, the median level was 106 ng/ml for light smokers (under 10 a day), 275 ng/ml for moderate smokers (10 to 19) and 344 ng/ml for heavy smokers (20 or more). The equivalent median figures for women were 81 ng/ml for light smokers, 262 ng/ml for moderate smokers and 319 ng/ml for heavy smokers.
Pipe and cigar smokers (among men) tended to have serum cotinine levels similar to those of light smokers if they were ex-cigarette smokers; otherwise, their levels were much lower than even light cigarette smokers. Its possible that the ex-cigarette smokers may have been more likely to inhale pipe or cigar smoke.
Serum cotinine levels were very similar for the other three categories of informants: ex-regular cigarette smokers, never regular cigarette smokers and never smoked at all. Table 4.12
‘Passive smoking’ refers to exposure to other people’s tobacco smoke. All informants who were not current cigarette smokers were asked if they were ever regularly exposed to other people’s tobacco smoke in any of the following places: in their own home; at work; in other peoples' homes; on public transport; in pubs; or in other public places.
The highest proportion of both sexes said they were most likely to be exposed to other people’s smoke in pubs than in any other places: 52% of men and 32% of women. Of course, exposure to smoke in pubs and other public places is relatively short-lived, whereas exposure in homes and workplaces is likely to be of much longer duration. The evidence suggests that quite high proportions of informants are exposed to other people’s smoke both in their workplaces (32% of men and 20% of women) and in their own homes (22% of men and 23% of women).
Overall, women were more likely than men to say they were not exposed to other people’s smoke in any of these places (31% compared with 22%).
Among both sexes, the likelihood of non-smokers saying they were not exposed to other people’s smoke increased with age. The proportions saying they had no exposure to smoke ranged from 7% of men and 14% of women aged 16-24 to 33% of men and 45% of women aged 55-64.
Among both men and women non-smokers, 16-24 year olds were significantly more likely to be exposed to smoke in their own home than any other age group: 30% of men, 33% of women (compared with averages of 22% and 23% among all men and women respectively).
Similarly, the proportion of those exposed to smoke at work also tended to decline with age. Among men, 36% of those aged 16-44 said they were exposed to smoke at work, compared with 23% of those aged 45-64. Among women, the figures were 23% and 15% respectively.
A similar trend was found for exposure to smoke in pubs, where the youngest age group were by far the most likely to mention such exposure: 64% of 16-24 year old men and 51% of women in this age group said they were exposed to smoke in pubs. This declined with age to 32% of men and 11% of women aged 55-64 Table 4.13
4.6.3 Passive smoking by social class
There were also significant variations in passive smoking by social class. The social class measure used is based on that of the chief income earner within the informant's household (see the Technical Report for details on the derivation of social class). Men in Social Class I and women in Social Classes I and II were by far the most likely to say they were not exposed to other people’s smoke in any of the places mentioned. Men and women in Social Class IV were the most likely to say they were exposed to other people’s smoke. Table 4.14
While these social class differences applied to all the types of places described, the largest differences between informants in non-manual and manual social classes were found for exposure to smoke in their own home, at their workplace, and in other people’s homes (with those in manual social classes being more likely than those in non-manual social classes to mention exposure to smoke in each of these places).
4.6.4 Passive smoking by level of serum cotinine
Among both male and female non-smokers, those who said they were exposed to smoke in their own home tended to have higher serum cotinine levels than those who were not: for example, for men non-smokers, only 9% of those who were exposed to smoke in the home had the lowest serum cotinine level (from 0 to 0.5 ng/ml) compared with 48% of those who were not exposed to smoke in the home. The equivalent figures for women were 13% and 54% respectively. Table 4.15
The results suggest that, while exposure to cigarette smoke in the home increases the concentration of serum cotinine in the blood, it does not, in general, raise it to levels associated with active cigarette smoking.
Table 4.16 shows the proportion of current cigarette smokers by social class (of the chief income earner within the informant’s household) within sex and age groups. The same data are also illustrated in Figure 4F. As this figure shows, informants in the manual social classes were much more likely than those in non-manual social classes to be current smokers. Table 4.16 and Figure 4F
While the sample sizes in some age/sex/social class groups were small, the differences in the proportions of current cigarette smokers between manual and non-manual social classes were found for all sex and age groups.
As can be seen from Table 4.17, informants in manual social classes were not only more likely to smoke than those in non-manual social classes, their average consumption of cigarettes was also higher. Among men, 24% of those in non-manual social classes reported that they were current cigarette smokers compared with 42% of men in manual social classes. Among male cigarette smokers, those from non-manual social classes smoked on average 16.7 cigarettes per day compared with 18.8 per day for those from manual social classes. Similar differences were found among women: 27% of women from non-manual social classes were current cigarette smokers compared with 44% of women from manual social classes; female cigarette smokers from non-manual social classes smoked on average 14.0 cigarettes per day compared with 16.5 per day for those from manual social classes. Table 4.17
Figure 4F: Proportion currently smoking cigarettes, by social class of chief income earner within age group
Among all informants, the proportion with serum cotinine levels of 20 ng/ml or more was 21% in non-manual social classes and 34% in manual social classes. (Table not shown.)
When self-reported cigarette smoking prevalence was taken as the dependent variable, a logistic regression showed significant social class effects for men and women separately after allowing for age. Again, for men and women separately, after age, there were significant social class effects when both the proportion of the sample smoking 20 or more cigarettes a day and the proportion with a serum cotinine level of 20 ng/ml or more were the dependent variables.
Among both men and women, self-reported cigarette smoking prevalence was highest in Greater Glasgow: 38% for men and 41% for women. The prevalence of current cigarette smoking was lowest in Lothian & Fife for men (31%) and in Highland & Islands for women (31%). In general, however, prevalence of cigarette smoking showed little variation between regions. Table 4.18
Looking at male current cigarette smokers, men in Borders, Dumfries & Galloway were the most likely to smoke 20 or more cigarettes a day (56%), while those in Highland & Islands (39%) were the least likely to smoke this amount. The mean number of cigarettes smoked per day was also highest in Borders, Dumfries & Galloway (21.0) and lowest in Highland & Islands (16.5).
Among current women smokers, there was less regional variation in level of cigarette consumption. On average, women in Greater Glasgow smoked the most per day (16.2), while those in the Highland & Islands smoked the least (14.9). Table 4.19
When misreporters (those who said that they were not current smokers but had serum cotinine levels of 20 ng/ml or more, as described in Section 4.5.3) were included, the proportions smoking at all (cigarettes, pipes or cigars) increased for both sexes in all regions: among men, the increase was between two and six percentage points, while among women it was between one and four percentage points. Table 4.20
A logistic regression showed no significant regional effects for either men or women, after allowing for age and social class, when self-reported cigarette smoking prevalence was the dependent variable. There were also no significant regional effects when the proportion smoking 20 or more cigarettes per day, or the proportion with a serum cotinine level of 20 ng/ml or more, were taken as the dependent variables. There were still no significant regional differences for any of the variables mentioned above when only age (rather than age and social class) was taken into account.
4.9 Comparison of Scotland and England
4.9.1 Self-reported smoking prevalence in Scotland and England
Data on smoking status and cigarette consumption from the Health Survey were compared with similar results from the 1995 Health Survey for England. There were significant differences between Scotland and England in the proportions of both sexes reporting that they currently smoked cigarettes. Men in Scotland were a bit more likely than those in England to be current cigarette smokers: 34% compared with 32%, a small, but statistically significant, difference. Among women, the difference was larger with 36% of women in Scotland classed as current cigarette smokers compared with 30% in England. Table 4.21
Figure 4G shows, separately for Scotland and England, and by age group, the proportion of men who reported smoking at all (cigarettes, pipe, cigar) and the proportion of men who reported smoking cigarettes.
Among men aged 16-45, there were no significant differences in the proportion of current cigarette smokers between the two countries. However, among those aged 45-64, a higher proportion of men reported smoking cigarettes in Scotland than in England: 34% compared with 27%. Figure 4G
With the inclusion of pipe and cigar smoking, there were still no significant differences between the younger age groups; but, significantly more men from older age groups in Scotland smoked at all compared with those in England: 43% of those aged 45-64 in Scotland, 35% in England.
Figure 4G: Proportion of men smoking in Scotland and England, by age
Figure 4H shows, separately for Scotland and England, by age group, the proportion of women who reported smoking cigarettes. While in England the proportion of women smoking cigarettes decreased with age, in Scotland there was no clear relationship between age and cigarette smoking. In England, more 16-24 year old women reported smoking cigarettes than any other age group, whereas in Scotland the largest age group reporting that they smoked cigarettes were women aged 25-34. For all age groups except the youngest, women in Scotland were more likely than women in England to be current cigarette smokers. Figure 4H
Figure 4H: Proportion of women smoking in Scotland and England, by age
4.9.2 Number of cigarettes smoked: comparison of Scotland and England
There were also significant differences between Scotland and England in the proportions of both sexes reporting that they smoked 20 or more cigarettes a day. Male smokers in Scotland were a bit more likely than those in England to report that they smoked 20 or more cigarettes a day (46% compared with 41%). For women the difference was larger, with 37% of female smokers in Scotland smoking 20 or more cigarettes a day compared with 29% in England.
Figure 4I shows the proportion of men smoking 20 or more cigarettes a day (based on current cigarette smokers) by age group, separately for Scotland and England. In both countries, the proportion of men smoking 20 or more cigarettes a day tended to increase with age (although there was a slight decrease in England between the last two age groups). In Scotland, however, the increase was greater than that in England: the increase in Scotland was from 23% in the youngest age group to 60% in the oldest, while in England, it was from 23% to 53% (in the 45-54 age group). Figure 4I
Figure 4I: Proportion of men smoking 20 or more cigarettes a day in Scotland and England, by age
Figure 4J shows the same data for women. In both Scotland and England, the proportion of women smoking more than 20 cigarettes a day increased with age until the 45-54 age group, and then decreased slightly among 55-64 year olds. However, in Scotland the increase was greater (from 18% in the youngest age group to 46% in the 45-54 age group) than that in England (from 16% to 37% in these groups). In all age groups, women in Scotland were more likely than those in England to smoke 20 or more cigarettes a day (although not all these differences were statistically significant). Figure 4J
Smokers in Scotland also had higher average daily cigarette consumption than those in England. Among men, the mean number of cigarettes smoked in Scotland was 18.1, while it was 15.9 in England. For all age groups, average daily consumption was greater in Scotland than in England. Similarly, among women, mean daily consumption was 15.4 in Scotland and 13.6 in England. For all age groups except the youngest, Scottish women smokers had significantly higher average daily cigarette consumption than those in England. Figure 4K
Figure 4J: Proportion of women smoking 20 or more cigarettes a day in Scotland and England, by age
Figure 4K: Average number of cigarettes per smoker per day, by age and country
4.9.3 Comparison of Scotland and Northern England
A comparison of smoking in Scotland with Northern England (which comprises the two regions of Northern & Yorkshire and North West) showed no significant difference between the proportion of men reporting that they currently smoked cigarettes: 34% of men in Scotland compared with 33% in Northern England. There was a significant difference for women, however, with women in Scotland a bit more likely to be current cigarette smokers than those in Northern England: 36% compared with 33%.
Figure 4L shows separately for men in Scotland and Northern England the proportion of self-reported current cigarette smokers and the average number of cigarettes smoked per day, by age group. In all age groups except the oldest, the proportion of male smokers was similar; among 54-64 year olds, a significantly higher proportion of men in Scotland than in Northern England were current smokers (34% compared with 24%). The number of cigarettes smoked per day was also similar for all age groups except the oldest. Among 55-64 year old male smokers, those in Scotland tended to smoke more cigarettes per day than those in Northern England (20.9 compared with 15.6). Figure 4L
Figure 4L: Proportion of men smoking and number of cigarettes per smoker per day in Scotland and Northern England, by age
Figure 4M shows the same data for women in Scotland and Northern England. As can be seen, the patterns were similar to those found among men. Thus, among younger women (16-45), there were few differences in the proportion classed as current smokers, but women in Scotland aged 45-64 were more likely than those in Northern England to say they currently smoked (36% compared with 30%). There were no significant differences between women in Scotland and Northern England in the average numbers of cigarettes smoked per day for any of the age groups. Figure 4M
Figure 4M: Proportion of women smoking and number of cigarettes per smoker per day in Scotland and Northern England, by age
4.10 Comparison of scottish health survey WITH THE 1984-86 scottish heart health study
The Scottish Heart Health Study (SHHS) looked at lifestyle and coronary heart disease risk factors in 10,359 men and women aged 40-59 between 1984 and 1986 (see Chapter 1 for a more complete description).9 One of the factors measured was cigarette smoking status. The SHHS found that 39% of men and 38% of women were, at that time, current cigarette smokers. A comparison of these figures with results from the Health Survey suggests that the proportion of the population (aged 40-59) classed as current cigarette smokers has decreased over the past decade, by about five percent in men (from 39% in SHHS to 34% in the Health Survey) and three percent in women (from 38% in SHHS to 35% in the Health Survey).10
Furthermore, the proportion of men and women who had never regularly smoked cigarettes increased over this period. At the time of the SHHS, 27% of men and 42% of women aged 40-59 had never regularly smoked cigarettes; the figures in the Health Survey were 36% for men and 45% for women in this age range.
Among men aged 40-59, the proportion of ex-regular cigarette smokers decreased during this time, from 34% in SHHS to 30% in the Health Survey. This reduction was not found among women, where the proportion of ex-regular cigarette smokers was 20% in both the SHHS and the Health Survey.
There was also a difference between the studies in the proportion of heavy smokers: in the SHHS, 25% of male smokers (aged 40-59) smoked 20 cigarettes or more a day compared with 20% in the Health Survey; the figures for women were 18% in the SHHS and 15% in the Health Survey.
It would appear therefore that, not only has the prevalence of cigarette smoking decreased over the past decade, but so has the prevalence of heavy smoking.