Tobacco smoking's changing trajectory in Australia

https://doi.org/10.1016/j.jbusres.2009.02.010Get rights and content

Abstract

This paper examines data from four Australian National Health Surveys and shows that decline in tobacco smoking prevalence experienced in Australia since World War II may not have continued as might have been inferred from the Lopez et al. model. The decline may have stabilized at approximately 20% of the adult population despite active anti-smoking campaigns. The data also suggest that changes in smoking status have not been consistent across all segments of the population despite all segments being exposed to the same fear-based campaign strategy. The data also suggests that while this health-scare approach may have coincided with improvements in the proportions of some groups of smokers who successfully quit the habit, an increase in the proportion of young females who have, at some stage, commenced to smoke has begun. If smoking prevalence is to be eradicated in Australia or other similar markets, campaigns must address cessation and prevention.

Introduction

The damage caused by tobacco smoking worldwide has been widely discussed in the literature for more than 50 years. Doll and Hill (1950) established the link between tobacco smoking and cancer in 1950. The United States Surgeon General declared tobacco smoking the single most important cause of preventable morbidity and mortality in his country, first in 1957 and regularly since then. (USDHHS, 1957, USDHHS, 1989, USDHHS, 2000) Both the Doll and the U.S. Surgeon General's findings have been regularly updated and confirmed in other countries ever since. Calculations of the cost of smoking to the community are approached from a number of different perspectives. Epidemiologists concluded that cigarettes bring on early death in as many as 50% of their lifelong users (Lopez et al., 1994, Peto et al., 1994). Peto et al. (2000) calculated that a global toll of 10 million deaths in the decades 1950 to 2000 could be directly attributable to tobacco smoking. In Australia, nearly 20,000 people die each year from smoking-related diseases out of a total population of not much more than 20 million (English et al., 1995).

Economists attempted to estimate the losses in productive capacity and other social costs to the community resulting from tobacco smoking. These losses are of the order of millions of quality-adjusted-life-years (QALYs) and many millions of dollars (Chaloupka and Warner, 1999, Collins and Lapsley, 1996, Collins and Lapsley, 1999a, Collins and Lapsley, 1999b, Collins and Lapsley, 2001, Single et al., 1998). An analysis of smokers' self-reports of expenditure on tobacco products in 1998–1999 indicates that approximately a $4 million of household expenditure on goods and services has been diverted into these harmful products. Furthermore, de Meyrick and Yusuf, (2006) suggest that this might be an underestimate by as much as 40%. Regardless of the approach taken, a general agreement that tobacco smoking exacts an enormous cost from the community exists.

Lopez et al. (1994) shows that tobacco smoking prevalence is a reliable predictor of tobacco-related illness and mortality. They used data from a large number of countries to show that smoking prevalence follows a four stage trajectory, with a corresponding, parallel trajectory for tobacco-related deaths, lagged by approximately 30 years. The trajectory resembles a typical product life cycle with an introductory, growth, maturity and decline stage (Levitt, 1965). Australia, along with the United Kingdom, United States, Western Europe and Canada, is shown to be well advanced in the fourth or decline stage. Corrao et al. (2000) suggest that any country can use this model to identify their position on the trajectory and then implement the strategies used by the stage four countries to reduce the time-scale and the amplitude of tobacco smoking's trajectory in their community. Although the Lopez model does not show smoking's trajectory beyond Stage VI, the end-point of the trajectory as interpreted by others such as Corrao is the complete cessation of smoking in the population and hence elimination of the damage caused by smoking.

This steady decline in smoking prevalence might be taken to indicate the sustained effectiveness of current anti-smoking initiatives in the Australian market. These initiatives are largely fear-inducing appeals, highlighting the damage that tobacco smoking does to a person's health. The strategy aims to communicate the cost to the community mentioned above, in a way that is relevant and persuasive to the population at large but which does not target any particular segment. An example of a magazine advertisement is attached (Appendix 1), featuring a still and a selection of copy from a current television commercial and is reproduced in graphic colour on billboards, posters and both the front and the back of cigarette packets. Note the graphic nature of the visual and the strong fear appeal in the copy.

Reliance on fear appeals has been discussed in the literature for some time. Authors from Rogers (1983) to Ruiter et al. (2003) and beyond provide guidance on factors to consider when developing such a campaign strategy but debate as to the effectiveness of these strategies, regardless of whether the guidelines continues. Rotfeld (2000) contains a review of contemporary issues. Hastings et al. (2004) provides a thorough discussion of the limitations of the strategy. All these papers advise against the sole reliance on this approach alone and on the continual increase in the level of fear engendered in the audience to overcome any audience resistance or wear-out in the message.

Section snippets

The national health survey

The principal data source used in this analysis is the Australian National Health Survey (NHS) (ABS Catalogue number 4363.0). Data from the last four surveys (1989/90, 1995, 2000/01 and 2004/5) are used in this analysis. Throughout the analysis, they are identified as the 1990, 1995, 2000 and 2005 surveys. The National Health Survey, as the name suggests, is a survey conducted by the Australian Bureau of Statistics (ABS) approximately every five years and covers an increasing range of

The effect of gender

These changes in the odds of having a particular smoking status have not been uniform across different groups in the community. For example, Fig. 3, Fig. 4 show the trends in the odds of belonging to one of the smoking status groups separately for males and for females.

Changes in the odds of belonging to one of the three smoking status groups in Fig. 3, Fig. 4 are approximately parallel with those observed in the overall sample (Fig. 2) but are more pronounced. The odds of a male being a

Discussion

The data indicates that rather than continuing a steady decline to the point of eradication, smoking prevalence may have levelled off in Australia at a little over 20% of the adult population. This may indicate that the projections implied by the Lopez model and the other projections mentioned above may not eventuate and that the avoidable damage that tobacco smoking causes the community will not be eradicated. This might be due to a wearing out in the effectiveness of the current fear appeal

Conclusion

The analysis indicates an urgent need to review the market segmentation and targeting strategies underpinning anti-smoking campaigns in Australia to identify the different needs and behaviours of different segments in the Australian population. Properly tailored campaigns can then be directed at these segments. The data also suggest an urgent need to examine the appropriateness of relying only on fear appeals when developing these new, more precisely targeted campaigns. This has international

References (22)

  • AndreasenA.R.

    Social Marketing in the 21st Century

    (2006)
  • ChaloupkaF.J. et al.

    The Economics of Smoking

  • CollinsD.J. et al.
  • CollinsD.J. et al.

    The Social Costs of Tobacco and the Social Benefits of Quit Victoria

    (1999)
  • CollinsD.J. et al.

    Human capital and demographic approaches to estimating the external costs of smoking

  • CollinsD.J. et al.

    The Social Costs of Tobacco in Western Australia and the Social Benefits of Reducing Western Australian Smoking Prevalence

    (2001)
  • de MeyrickJ.H. et al.

    The application of household expenditure data in the development of anti-smoking campaigns

    J Health Educ

    (2006)
  • DollR. et al.

    Smoking and carcinoma of the lung: a preliminary report

    BMJ

    (1950)
  • EnglishD. et al.
  • HastingsG. et al.

    Fear appeals in social marketing strategic and ethical reasons for concern

    Psychol Mark

    (2004)
  • The author wishes to thank the anonymous reviewers whose helpful suggestions greatly improved this paper.

    View full text