| Health Survey for England - The Health of Minority Ethnic Groups '99 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
4 Use of tobacco products Richard Boreham
4.1.1 Background The White Paper Saving Lives: Our Healthier Nation1 identifies smoking as a major risk factor for deaths from cancer and coronary heart disease and stroke - two of the four health targets set. In addition the importance attached by the government to reductions in levels of smoking in all social classes is emphasised in the Smoking Kills: A White Paper on Tobacco.2 4.1.2 Chapter content The Health Survey collects data about use of tobacco by means of both self-reported behaviour and measurements of the cotinine levels in informants' saliva. The chapter looks at cigarette smoking prevalence, use of chewing tobacco products, use of any tobacco products (including cigarettes, cigars, pipes and chewing tobacco), and estimated overall tobacco use based both on reported usage and on cotinine levels. Levels of smoking are measured in two ways: self-reported number of cigarettes smoked, and the prevalence of high levels of saliva cotinine. 4.1.3 Questions asked about smoking and use of other tobacco products in the Health Surveys In 1999, as in previous surveys, information about smoking was collected from those aged 16 and 17 by means of a self-completion questionnaire, while for those aged 18 or over3 it was collected as part of the main interview. Questions about chewing tobacco were introduced in 1999, and were asked of South Asian groups only, as these were believed to be the only groups in which the use of chewing tobacco was likely to be significant. The reason for including questions on chewing tobacco in the Health Survey was the associated health risk. Some oral tobacco products (such as snuff) seem to have rather little risk attached to them, but chewing tobacco is a major risk factor for oral cancer, which is particularly prevalent in the Indian subcontinent. Chewing tobacco comes in three forms: plain chewing tobacco, tobacco paste (zarda) and paan masala (tobacco mixed with betel nut). Tobacco paste is generally eaten wrapped in paan (betel leaf). Use of any of these is defined as use of chewing tobacco. In addition informants were asked whether they used hukka (tobacco smoked through water using a pipe), bidi (rolled tobacco leaf) and paan without tobacco (plain betel leaf). Use of hukka or bidi was included in the definition of use of any tobacco products, but not of chewing tobacco. Paan without tobacco was not included in any definition of use of tobacco products. 4.1.4 Cotinine Cotinine is a metabolite of nicotine. It is one of several biological markers that are indicators of tobacco use (others include carbon monoxide and thiocyanate) and is generally considered the most useful. It can be measured in, among other things, saliva or serum. Cotinine has a half-life in the body of between 16 and 20 hours, which means that it will detect regular tobacco use but will not detect occasional tobacco use if the last occasion was several days ago. Before 1998, cotinine levels in the Health Survey were measured in serum in adults, but from 1998 were measured in saliva, primarily to increase the number of people being measured as more people refuse to give a blood sample than a saliva sample. Cotinine levels measured in saliva are higher than those measured in serum but there is a linear relationship between the two measurements.4 In the Health Survey, cotinine samples are collected by the nurse. In the 1999 survey, nurse visits were confined to minority ethnic group members, so there is no cotinine data for the general population. Comparative general population data have therefore been taken from the 1998 survey. 4.2.1 Introduction Informants were asked whether they had ever smoked a cigarette, cigar or pipe, and then were asked whether they smoked cigarettes nowadays, and, if they didn't, whether they had ever smoked cigarettes regularly. The answers to these questions were combined to create a cigarette smoking classification consisting of current cigarette smoker, ex-regular cigarette smoker and never regular cigarette smoker. Pipe and cigar smoking is disregarded in this classification, but pipe and cigar smokers are included in the definition of users of any tobacco products (along with users of chewing tobacco). 4.2.2 Cigarette smoking, by minority ethnic group Cigarette smoking among men Among the general population 27% of men said they were current smokers, 31%
used to smoke regularly and 42% had never smoked regularly. The inset
table shows the proportion reporting current cigarette smoking in each
minority ethnic group, together with the age-adjusted risk ratio relative
to men in the general population. The risk ratios are also shown in
Figure 4A (on which the vertical axis is logarithmic).
Bangladeshi and Irish men were the groups with the highest self-reported prevalence
of current cigarette smoking (Bangladeshi men age-standardised risk
ratio relative to men in the general population 1.57, Irish men risk
ratio 1.43). Black Caribbean men were the only other group who were
more likely to smoke cigarettes than men in the general population (risk
ratio 1.26). Indian and Chinese men were less likely to smoke cigarettes
then men in the general population (Indian men risk ratio 0.78; Chinese
men risk ratio 0.62). In the general population around half (54%) of men who had ever smoked
regularly were not current smokers. Among all minority ethnic groups,
other than Chinese men, men who had ever smoked regularly were less
likely than men in the general population to have stopped smoking. Pakistani
and Bangladeshi men were the least likely groups of men to have stopped
smoking, with only around one in five men who had ever smoked regularly
having given up.
In the general population the prevalence of cigarette smoking among men decreased with age. Men in minority ethnic groups broadly followed a similar pattern, apart from Bangladeshi men, whose cigarette smoking prevalence actually increased with age, and also Black Caribbean men, where cigarette smoking was most prevalent in those aged 35-54. Table 4.2, Figure 4B Cigarette smoking among women
Among the general population 27% of women were current smokers, 22% used to smoke regularly and 52% had never smoked regularly.
Irish women were the only group who were more likely to be current cigarette smokers than women in the general population (risk ratio 1.16). Black Caribbean women were marginally less likely than women generally to be current cigarette smokers (risk ratio 0.85). Cigarette smoking prevalence was very low among South Asian and Chinese women,
particularly Bangladeshi women. Risk ratios for current cigarette smoking
were 0.07 for Bangladeshi women, 0.14 for Pakistani women, 0.19 for
Indian women and 0.31 for Chinese women. Women in the general population who had ever smoked cigarettes regularly were less likely than men to have stopped being current cigarette smokers, although this gender difference may be due to men switching to cigars or pipes.5 45% of women who had ever smoked regularly no longer smoked cigarettes, compared with 54% of men. The proportion of Black Caribbean women (30%) who had stopped smoking after smoking regularly was lower than the 45% for women in the general population, while the proportion of Irish women who had stopped smoking cigarettes (57%) was similar to that of women in the general population. The numbers of South Asian and Chinese women who had ever smoked cigarettes regularly were so small that it is not possible to compare the proportions who were still smoking with other groups. As with men, the prevalence of cigarette smoking among women decreased
with age in the general population, and among Black Caribbean, Indian,
Chinese and Irish women. Among Pakistani and Bangladeshi women the overall
prevalence of cigarette smoking was too low for any clear pattern of
variation by age to be seen. 4.2.3 Socio-economic variations in self-reported cigarette smoking Among men in the general population both social class of head of household
and equivalised household income were associated with current cigarette
smoking, the prevalence of smoking cigarettes being higher in the lowest
household income tertile and among manual social classes. In the general
population, cigarette smoking prevalence was 34% among men in manual
households compared with 19% of men in non-manual households. There
was a similar pattern among Black Caribbean, Bangladeshi and Irish men
(among Black Caribbean men 38% in manual households smoked compared
with 25% in non-manual households, the equivalent figures for Bangladeshi
men being 47% and 33% and for Irish men 44% and 31%), but there was
no clear association between social class or household income and cigarette
smoking for Indian, Pakistani or Chinese men.
The relationship of social class and equivalised household income
to cigarette smoking was the same for women as for men in the general
population (for example, in the general population, cigarette smoking
prevalence was 32% among women in manual households compared with 21%
in non-manual households), but not in all minority ethnic groups. Among
South Asian women, the social class gradient was in the opposite direction,
with cigarette smoking being more prevalent in non-manual women, and
there was no clear household income gradient. Cigarette smoking prevalence
among Chinese women did not show a social class gradient, but did show
an income gradient; 14% of Chinese women in the top household income
tertile smoked cigarettes compared with 5% in the bottom household income
tertile. Irish women in manual social classes and the lowest income
households were more likely to smoke cigarettes. There were no relationships
between cigarette smoking and either social class or household income
among Black Caribbean women. 4.3 Number of cigarettes smoked 4.3.1 Introduction Informants who said they smoked cigarettes were asked how many cigarettes they normally smoked per day on weekdays, and how many they smoked at weekends. From this the total number of cigarettes per week was estimated, and divided by seven to obtain the average number of cigarettes per day. It is possible to analyse levels of smoking in a number of ways. The number of cigarettes smoked can either be grouped into categories (normally less than 10, 10 or more but less than 20, 20 or more) or can be expressed as a mean, and can be based on all informants or only on those who smoke cigarettes. The method of analysis chosen for this section is to look at the banded number of cigarettes smoked based on current cigarette smokers. 4.3.2 Number of cigarettes smoked, by minority ethnic group Figure 4E shows the relative risk of smoking 20 or more cigarettes per day based on current cigarette smokers. People smoking 20 or more cigarettes per day are classified as 'heavy smokers'. Among men smokers in the general population, 37% were heavy smokers. Men in
all minority ethnic groups, other than Irish men, who smoked were less
likely to be heavy smokers than men in the general population. Women smokers were less likely to be heavy smokers than their male counterparts; 27% of women smokers smoked 20 or more cigarettes per day compared with 37% of men smokers. Irish women who smoked were as likely to be heavy smokers as women in the general population, but Black Caribbean women smokers were less likely to be heavy smokers. It also appears that Indian, Pakistani and Chinese women who smoked were less likely to be heavy smokers, although these figures should be treated with caution because of the small numbers of women in these minority ethnic groups who were current cigarette smokers. 4.4.1 Introduction South Asian informants were asked about their use of other tobacco products, namely use of chewing tobacco, or tobacco in conjunction with other products such as paan, and also use of paan without tobacco. Only observed percentages are presented in this section, as there are no general population figures to use for comparative purposes. 4.4.2 Use of chewing tobacco, by South Asian minority ethnic groups Informants were classified as using chewing tobacco if they used either paan
with tobacco, paan masala or chewing tobacco. The inset table shows
the proportion of men and of women, within each of the South Asian groups,
who used any form of chewing tobacco.
Prevalence was highest among Bangladeshi men and women. Paan with tobacco was the most commonly used tobacco product, used by 14% of Bangladeshi men and 23% of Bangladeshi women. Paan without tobacco was also widely used, by 28% of Bangladeshi men and 22%
of Bangladeshi women. The prevalence of use of chewing tobacco among Bangladeshi men and women increased with age. 14% of Bangladeshi men aged 16-34 used chewing tobacco, compared with 23% of Bangladeshi men aged 35-54 and 28% of Bangladeshi men aged 55 and over. Equivalent prevalence figures for Bangladeshi women were 15% among women aged 16-34, 43% among women aged 35-54 and 56% among women aged 55 and over. Among Bangladeshi men, chewing tobacco was more likely to be used in conjunction
with cigarettes than instead of cigarettes. 12% of Bangladeshi men used
chewing tobacco and smoked cigarettes, while 8% used chewing tobacco,
but did not report smoking cigarettes. Chewing tobacco was the main
tobacco product of Bangladeshi women, of whom 26% used chewing tobacco
and did not smoke cigarettes, while less than 1% chewed tobacco and
smoked cigarettes. 4.4.3 Socio-economic variations in use of chewing tobacco Bangladeshi men in households in manual social classes were more likely to use chewing tobacco (22%) than Bangladeshi men in non-manual households (12%). Similar patterns were observed for Bangladeshi women: 17% of Bangladeshi women in non-manual households used chewing tobacco compared with 26% of Bangladeshi women in manual households. There were no clear relationships between household income and use of chewing tobacco among Bangladeshi men and women. There were no clear relationships between use of chewing tobacco and
either social class or household income for Indian and Pakistani men
and women. 4.5 Prevalence and cotinine-adjusted prevalence of use of tobacco products 4.5.1 Introduction This section looks at self-reported overall use of any tobacco products, including cigarettes, cigars, pipes and chewing tobacco. A saliva cotinine level of 15 ng/ml or over can be taken as an indicator of current tobacco use. An adjusted measurement of the use of tobacco products was created by defining tobacco users who either reported use or had saliva cotinine of 15 ng/ml or over. Potential under-reporting can be assessed by comparing this measure with self-report on its own. Data were given an additional weight to correct for the response bias in obtaining a valid saliva sample.6 Table 4.11 shows the distribution before this additional weight was applied. 4.5.2 Prevalence of use of tobacco products, by minority ethnic group The use of tobacco products other than cigarettes varied between groups, and the extent to which overall tobacco prevalence exceeded cigarette smoking prevalence also varied. Among men, the difference was greatest in the case of Bangladeshis, of whom 44% smoked cigarettes but a total of 53% used any tobacco product. After cigarettes, by far the most common form of tobacco used by Bangladeshi men was chewing tobacco. Indian men used a wider variety of tobacco products (cigars 9%, pipes 4%, chewing tobacco 6%) and overall tobacco use was 28%, compared with a cigarette smoking prevalence of 23%. Among other groups, the predominant form of tobacco use other than cigarette smoking was cigar smoking, and overall tobacco use was 3% to 5% higher than cigarette smoking prevalence. Bangladeshi women had a quite different pattern of use from other
women, with almost all their reported tobacco use being chewing tobacco.
Cigarettes were the main tobacco product used by women in other groups,
and overall prevalence of use of tobacco products was similar to prevalence
of cigarette smoking. 4.5.3 Prevalence of cotinine-adjusted use of tobacco products, by minority ethnic group A saliva cotinine level of 15 ng/ml or over can be taken as an indicator of
tobacco use, and an adjusted tobacco use prevalence can be created by
defining people as tobacco users if they either report using a tobacco
product, or have a saliva cotinine level of 15 ng/ml or over. The inset
table compares this adjusted use prevalence estimate with the estimate
derived solely from self-report. In all cases, the adjusted estimate
was higher. The table shows the magnitude of the difference between
them.
a As noted in Section 4.1.4, cotinine was not measured for the general population in 1999, and comparative data Among men, the difference was greatest in the case of Pakistani men. 28% of Pakistani men reported using any tobacco products, but an additional 9% had cotinine levels that suggested regular use of tobacco. The difference was also greater for Black Caribbean, Indian and Bangladeshi men than for men in the general population. Among women, Bangladeshi women showed the largest difference between the self-report
and adjusted estimates. 27% of Bangladeshi women reported using tobacco
products, but a further 12% had cotinine levels of 15 ng/ml or over.
Indian and Pakistani women showed differences that were smaller in absolute
terms, but (because of low tobacco use prevalence in these groups) were
large in relative terms, adjusted prevalence being at least half again
as large as unadjusted prevalence. 8% of Indian women and 7% of Pakistani
women reported using tobacco, but for both groups the adjusted overall
use estimate was 12%. References and notes 1 Saving Lives: Our Healthier Nation. The Stationery Office, London, 1999. 2 Smoking Kills: A White Paper on Tobacco. The Stationery Office, London, 1998. 3 A small number of persons aged 18-25 also answered these questions by self-completion rather than by interview, as the interviewer judged that more reliable responses might be obtained by this method. 4 Hedges BM, Jarvis MJ. Cigarette smoking (Chapter 6, Section 6.4.2) in Prescott-Clarke P, Primatesta P (eds) Health Survey for England: The Health of Young People '95-97. The Stationery Office, London, 1998. 5 Jarvis MJ. (1994) Gender differences in smoking cessation: real or myth? Tobacco Control 1994; 3:324-328. 6 Informants who had a valid saliva cotinine
measurement had different tobacco use characteristics from all informants,
and this response bias was different for different minority ethnic groups.
For the analysis of saliva cotinine, informants were given an additional
weight so that within minority ethnic group and sex, the profile of
self-reported tobacco use among those with a saliva cotinine measurement
matched the profile of tobacco use among all informants. 4.1 Self-reported cigarette smoking status, by minority ethnic group 4.2 Self-reported cigarette smoking status, by age within minority ethnic group 4.3 Self-reported cigarette smoking status, by social class of head of household within minority ethnic group 4.4 Self-reported cigarette smoking status, by equivalised household income tertile within minority ethnic group 4.5 Number of cigarettes smoked, by minority ethnic group 4.6 Types of substances used, by South Asian minority ethnic group 4.7 Prevalence of chewing tobacco, by age within South Asian minority ethnic group 4.8 Prevalence of chewing tobacco, by social class of head of household within South Asian minority ethnic group 4.9 Prevalence of chewing tobacco, by equivalised household income tertile within South Asian minority ethnic group 4.10 Self-reported current use of tobacco products, by minority ethnic group 4.11 Comparison of saliva cotinine sample with total sample, by minority ethnic group 4.12 Prevalence of saliva cotinine 15 ng/ml or over, by minority ethnic group 4.13 Prevalence of saliva cotinine 15 ng/ml or over, by age within minority ethnic group 4.14 Prevalence of saliva cotinine 15 ng/ml or over, by social class of head of household within minority ethnic group 4.15 Prevalence of saliva cotinine 15 ng/ml or over, by equivalised household income tertile within minority ethnic group 4.16 Prevalence of self-reported use of tobacco products, of saliva cotinine of 15 ng/ml or over, and of adjusted tobacco use, by minority ethnic group
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
