1.3 Prevalence of smoking—adults

Show / hide chapter menu

Last updated: February 2019      Suggested citation: Greenhalgh, EM, Bayly, M, & Winstanley, MH. 1.3 Prevalence of smoking—adults. In Scollo, MM and Winstanley, MH [editors]. Tobacco in Australia: Facts and issues. Melbourne: Cancer Council Victoria; 2019. Available from http://www.tobaccoinaustralia.org.au/chapter-1-prevalence/1-3-prevalence-of-smoking-adults

1.3.1 Latest estimates of smoking prevalence in Australia

Recent data on  prevalence of smoking in Australia can be found in the latest report of the 2016 National Drug Strategy Household Survey published by the Australian Institute of Health and Welfare,¹ and the 2017–18 National Health Survey, published by the Australian Bureau of Statistics.²

National Drug Strategy Household Survey

The National Drug Strategy Household Survey most commonly refers to daily smoking rates among those 14 years and over. However, also included are figures for those 18 years and over who smoke daily, at least weekly and less than weekly. These figures are reproduced in Table 1.3.1.

In shorthand speech, the 'prevalence of smoking among adults in Australia' could be cited simply as 14.2%.

National Health Survey

According to the Australian Bureau of Statistics' National Health Survey, the prevalence of daily smoking for Australians aged 18 and over in 2017–18 was 13.8%, and 13.3% among people aged 15 years and over.²

1.3.2 Historical trends in smoking prevalence in Australia

Measurements of the prevalence of smoking in Australia first became available in 1945. Limited survey data³ are available for the years between then and 1974, when the Anti-Cancer Council of Victoria (now Cancer Council Victoria) conducted its first national survey.^4,5 These early data show that in around the middle of the last century, a clear majority of males aged 16 and over were smokers, compared to about one-quarter of females (Table 1.3.2). In the following decades smoking among men declined, probably in response to the initial publicity regarding the health effects of smoking, which first emerged in the 1950s and early 1960s.^6-9 Women have always had a lower prevalence of smoking than men, but smoking among women continued to increase in the 1970s.

The findings of the early studies from Cancer Council Victoria are broadly confirmed by those of a survey by the Australian Bureau of Statistics undertaken in 1977, which found that 36% of the adult population (aged 18 and over) were smokers: 43% of men and 29% of women.¹⁰

Table 1.3.3 shows the proportion of smokers in the population aged 18 and over from 1980 to 1998. Taking into account the ageing of the population and other demographic trends over that period by standardising this data to the 2001 population structure, the prevalence of smoking declined for both sexes over this period, the most dramatic drop occurring among males between 1983 and 1986, when prevalence decreased relatively by 15%. The differential in smoking rates between the sexes also continued to close (while remaining statistically significant across the years to 1998), largely due to greater numbers of men quitting smoking during the mid-to-late 1980s. However, the overall rate of decline seen during the 1980s did not continue into the 1990s, where the prevalence of smoking levelled at about 27%.

After this relatively static period in the 1990s, the decline resumed for both sexes after 1998. Data from the National Drug Strategy household Survey shows that since 2001, there has been a statistically significant decline in the prevalence of smoking for men and women(controlling for age), and the total sample (controlling for age and gender) (see Table 1.3.4).

Hill and colleagues have suggested that the pattern of decline in smoking prevalence correlates with the level of tobacco control activities occurring at the time.^11,12 The drop in male smoking rates seen in the early 1980s coincided with a period of new, well-funded media-led Quit campaigns¹¹ and an upsurge in debate about tobacco control issues in the media, fuelled by the outspoken campaigning of groups such as the Australian Council on Smoking and Health and Action on Smoking and Health, and the widely publicised activities of the fringe groups MOP UP and BUGA UP.ⁱ Conversely, the steady prevalence rates in both sexes seen during the 1990s correspond with a lull in legislative activity concerning tobacco advertising and smoking restrictions, and also with a sharp reduction in per capita expenditure on public education campaigns.¹¹

The subsequent downturn in smoking prevalence seen by the end of the 1990s—see figure 1.3.1—may be attributable to the combined effects of increased tobacco taxes,¹³ additional smokefree legislation, and the National Tobacco Campaign, a mass-media led program aimed at encouraging cessation, which was launched in June 1997¹⁴ and ran over several subsequent years.^15,16  (See also Chapter 10.)

Data from the ABS National Health Survey² show that since 2001, the proportion of adults who are daily smokers has decreased from 22.3% to 13.8% in 2017-18. Over recent years however, the daily smoking rate has remained relatively stable (14.5% in 2014-15)—see Figure 1.3.2.

Figure 1.3.2 Proportion of Australians 18+ who smoke at least daily—2001 to 2017–18

Source: Table 1: ABS 4364.0.55.001 - National Health Survey: First Results, 2017-18  

Another long-running source of information about the prevalence of smoking is the Single Source Survey established in 1974 by the Roy Morgan Research Company. Door-to-door interviews conducted each week throughout Australia generate data from over 50,000 survey respondents each year.

Cancer Council Victoria analysis of monthly data from this survey from the largest five-capital cities over the 15 years to August 2017 shows a continuing decline in smoking since mid-2006—see Figure 1.3.3.

Figure 1.3.3 Proportions of Australians 18+ living in largest five capital cities who report being current smokers—12-month rolling average, January 2001 to August 2017

Source: Cancer Council Victoria calculation of 12-month rolling average, Roy Morgan Research Single Source survey data, five largest capital cities, (factory-made cigarettes, RYO, cigars and pipe tobacco), January 2001 to August 2017 (Wakefield et al IMPACT study in progress).

Smoking remains a leading cause of death and disease in Australia, killing an estimated 18,762 people annually.¹⁷ Since about one in six men and about one in eight women still smoke,¹ and given that two–thirds of these smokers can be expected to die because of their tobacco use if they do not quit,¹⁸ the sequelae of tobacco-caused death and disease will remain for decades to come. Mortality caused by tobacco use is discussed in Chapter 3.

1.3.3 International comparisons of adult smoking prevalence

An estimated 1.25 billion adults worldwide are smokers.¹⁹ In general, the prevalence of smoking is declining in industrialised countries in Northern and Western Europe, North America and the Western Pacific region, and increasing in some countries in Asia, South America and Africa.¹⁹ As global patterns in tobacco use change, the burden of death can be expected to shift dramatically from the developed world to less wealthy countries.²⁰ More than two-thirds of smoking-related deaths occur in low- and middle-income countries,²¹ where the large majority of the world’s smokers now live. This is at least in part due to a lack of adequate tobacco controls.²²  

A paradigm illustrating the typical progression of tobacco use worldwide, first proposed by Lopez and colleagues²³ and later adapted by the WHO,²⁴ is reproduced in Figure 1.3.4. Many (but not all) countries’ experiences of patterns of tobacco use fit this model. 

  

Stage I of the model is marked by a low smoking prevalence (below 20%), generally limited to males and accompanied by little evident increase in tobacco-caused chronic illness. Countries at stage I have not yet become major consumers in the global tobacco economy, but represent untapped potential for the tobacco industry. Some countries in sub-Saharan Africa fit into this stage in the model.²⁴ The importance of tobacco farming in some countries in the region (e.g. Zimbabwe and Malawi) may act as a deterrent to the introduction of tobacco control policies. Zimbabwe is among the largest producers of tobacco in the world and is a major exporter; concerns about the health consequences of tobacco use are not high on the national agenda.²⁵   

In stage II of the paradigm, male prevalence of smoking has soared to more than 50% in men, and women’s smoking rates are now increasing. Uptake of smoking is occurring at an earlier age, and, although there is now evidence of increased lung cancer and other chronic illness due to smoking among men, public and political understanding of and support for tobacco control initiatives is still not widespread. Countries that fit into this transitional stage include Japan, some nations within the Southeast Asian, Latin American and North African regions, and to a lesser extent, China. (The case of China is discussed further below).²⁴  

Stage III of the epidemic has been reached when smoking prevalence peaks and begins to decline in both sexes, although deaths caused by smoking continue to increase because of earlier high smoking rates. Health education programs are better developed, and smoking becomes less accepted among the more educated groups of society. Smoking becomes less socially acceptable and the climate is increasingly conducive to the introduction of tobacco control policies. Certain countries within Eastern and Southern Europe and Latin America are at this point on the continuum.²⁴

Evolution into stage IV is marked by a continued distinct but gradual downturn in smoking prevalence among both males and females. Male deaths from smoking begin to decline, but female death rates continue to rise, reflecting earlier smoking patterns. Parts of Western Europe, the UK, the US, Canada, New Zealand and Australia are at various points on the continuum in the fourth stage of the tobacco epidemic.²⁴ However, comprehensive and continually monitored public health strategies remain critical to maintain and reinforce declines in smoking prevalence.²⁴

As noted above, there are some countries for which the paradigm devised by Lopez and colleagues in Figure 1.3.1 does not fit. This is especially so in nations in which female smoking rates have not shown a pattern of steady increase in stage II, despite high prevalence among males, most likely due to social or cultural constraints. For example, men in China and Indonesia have maintained high rates of smoking for many years, while female prevalence has remained in single digits. Nevertheless the WHO model described above does provide a useful framework into which many countries can be placed, and may enable countries currently at an earlier stage in the paradigm to recognise their situation, learn from international experience and introduce strong public health measures that will reduce the impact of tobacco on their population. Singapore provides a successful example of early intervention. In the early 1970s, while at stage II of the model, the Singaporean government initiated a series of tobacco control measures which capped smoking prevalence at a relatively low level, effectively averting the later stages of the epidemic. Thousands of tobacco-caused deaths in Singapore have been prevented as a result of this early, decisive action.²³  

Table 1.3.5 presents statistics on smoking prevalence from a number of different countries. 

Figure1.3.5 and Figure 1.3.6 show prevalence data collected by the Organisation for Economic Co-operation and Development (OECD) of its member countries,²⁶ and the data on prevalence on less developed countries in Table 1.3.2 are taken from a wide variety of sources compiled for The Tobacco Atlas (Third Edition) published by the American Cancer Society and World Lung Foundation.²¹ These tables are provided in order to give a general global overview. It is important to note that data sets between countries are not directly comparable, due to differences in sampling (most crucially the year of the survey) and definitions (daily, regular (daily plus weekly) or current (daily, weekly or less than weekly) smokers), and that overall prevalence figures such as those provided by the tables may mask higher smoking levels among particular sub-groups of the population (most notably men in comparison to women). Further, studies which only take into account the smoking of manufactured cigarettes will underestimate tobacco use in countries where tobacco is widely used in other forms, such as in pipes, hand-rolled leaves or as chewing tobacco. This is a key consideration in countries where alternative methods of tobacco use are prevalent, such as in Sweden²⁷ and throughout much of Southern and Southeast Asia.^{28, 29} The interested reader is referred in the first instance to the primary sources, which explain the parameters of each study.