Between 1950 and 2008, an estimated 900,000 Australians died prematurely because they smoked. It is expected that the toll will exceed one million within the next decade. Despite reductions in the prevalence of tobacco use in the last 30 years, smoking is still the largest preventable cause of death and disease in the country with approximately 19 per cent of the population smoking daily in 2007–8 (over three million people). Tobacco smoking costs the Australian health care system more than $30 billion a year. Smoking also causes numerous health conditions, including certain ocular diseases that do not result in death but do greatly impact quality of life.[4-6] Smoking causes or contributes to approximately 20 per cent of all new cases of blindness in Australia for people over the age of 50. The most common eye disease causally linked to smoking is age-related macular degeneration (AMD). AMD is the leading primary cause of blindness in Australia with an estimated 150,000 Australians having AMD as of 2004 and approximately 17,700 new cases of wet macular degeneration, reported annually.
Perceived health risk plays an important role in initiating and quitting smoking. Smokers with a high knowledge of the health effects of smoking are more likely to quit and remain abstinent.[12, 13] Australia is one of only a few jurisdictions in the world, where public and non-governmental sectors are working toward increasing public knowledge about the risk of smoking on ocular health. Australia's National Tobacco Campaign, launched in 1997, has included numerous health promotion initiatives, including a television advertisement that linked smoking and ‘loss of eye sight’ starting in 1998. The advertisement explained how smoking can damage blood vessels in the back of the eye, leading to visual loss. The ‘eye’ advertisement has been run intermittently since 1998 and is credited with encouraging more Australian smokers to call the national Quitline than any other message about the health effects from smoking. This suggests that the message that ‘smoking causes blindness’ is effective at increasing interest in cessation.
A 2011 study reported that a low proportion of adult smokers from Canada (13 per cent), the United States (10 per cent) and the United Kingdom (10 per cent) believed that smoking can cause blindness. In contrast, 47 per cent of Australian daily smokers believed that smoking causes blindness. The study noted that Australia was the only country during the sampling period to have national social marketing campaigns about smoking and its effects on ocular health. This finding suggests that social marketing campaigns likely contributed to higher knowledge among smokers about the causal links between smoking and blindness; however, even with years of government-sponsored social marketing, less than half of the Australian smokers reported that they were aware of the connection between smoking and blindness.
The Australian Commonwealth government has required tobacco manufacturers to include textual health warnings on cigarette packages since 1973. In March 2006, new pictorial health warnings (pictures and written warnings) were required on most domestic and imported tobacco product packaging. The intended purpose of these warning labels was to increase knowledge of the health effects relating to smoking, encourage smoking cessation and discourage smoking uptake or relapse. One of the 14 warning labels, commonly referred to as the ‘eye’, included the written warning, ‘Smoking causes blindness’. The graphic used was a still from the 1998 television advertisement. When these graphic warning labels were evaluated in discussion groups with smokers, ex-smokers and non-smokers for the Australian Department of Health and Ageing, ‘the eye’ image dominated responses. The image was described as ‘grotesque’ and was considered to have high emotional impact (Figure 1).
The current study sought to examine how the introduction of the ‘Smoking causes blindness’ warning label may have impacted smokers' knowledge of this health effect in Australia and to compare this with the knowledge of smokers in comparable countries, where there were no similar warning labels on cigarette packs. This study also sought to understand if personal characteristics, such as sex or socio-economic status, were related to knowledge about smoking causing blindness. Data used in this study were collected between 2004 and 2011, with sampling taking place both before and after the introduction of the Australian warning label.
The study analysed data from the International Tobacco Control Policy Evaluation Project (ITC) Four Country Survey conducted in Australia (AU), Canada (CA), United Kingdom (UK) and United States (US). Since 2002, the ITC Four Country Survey has collected data from nationally representative samples of adult smokers (18 years and older), approximately annually. Since 2004, Wave 3, the survey has included a question to assess smokers' knowledge of blindness and smoking. During the present study period (2004–2011), there was no national social marketing campaign that explained the causal link between smoking and ‘blindness’ in Canada, UK or the US. In 2003 and 2008, the UK introduced new warning labels; however, only Australia had a tobacco product warning label that communicated health knowledge about smoking and ocular health.
- Top of page
- Grants and Financial Assistance
The sample of smokers from the pre-introduction of the ‘eye’ warning label on cigarette packages in Australia (June 2004 to February 2007), included 13,143 respondents (Australia: n = 3,151, Canada: n = 3,221, UK: n = 3,196 and US: n = 3,575,). The sample of smokers from the post-introduction of the ‘eye’ warning label surveyed September 2007 to May 2011 included 4,329 respondents (Australia: n = 884, Canada: n = 1,083, UK: n = 893 and US: n = 1,469). Full sample characteristics for the pre/post-introduction of the warning label can be found in Tables 1A and 1B.
Figure 2 presents country-specific changes in proportions of smokers reporting that smoking causes blindness. Proportions are given for both before and after introduction of the ‘eye’ warning label in Australia. Prior to the introduction of the warning label, less than half of Australian smokers (48.8 per cent, 95% CI: 46.8, 50.8) reported that they believed that smoking causes blindness. This proportion significantly increased (χ2 = 43.04, p < 0.001) to 62.4 per cent (95% CI: 58.6, 66.0) after the warning labels were introduced in March 2007.
Figure 2. Changes in reported knowledge of the causal association between smoking and ‘blindness’, by country (2004/2006–2007/2011)
*Indicates country-specific significant change in proportion between 2004/2006 and 2007/2011 using χ2 test α = 0.05 significance level
Download figure to PowerPoint
In Canada, UK and the US, the proportions of adult smokers reporting the belief that smoking causes blindness were significantly lower compared with both Australian samples before and after the introduction of the Australian warning label. In Canada and the US, there was no significant change in proportions (p > 0.05 for χ2 tests) between pre- and post-values, with knowledge topping out at 16.1 per cent (95% CI: 13.4, 19.2) and 9.8 per cent (95% CI: 7.8, 12.2), respectively, in the post period (2007 to 2011). In the UK, the proportion of adult smokers reporting that smoking causes blindness significantly increased (χ2 = 61.67, p < 0.001) from approximately 9.9 per cent (95% CI: 8.8, 11.3) to 17.9 per cent (95% CI: 15.8, 21.9) between the pre- and post-study periods. Looking at the proportion of UK smokers reporting knowledge of blindness and smoking year over year, a significant change took place between the sampling period of June–December 2004 (Wave 3) and October 2005–January 2006 (Wave 4); knowledge went from 6.9 per cent (95% CI: 5.7, 8.5) to 16.1 per cent (95% CI: 12.8, 20.0). Knowledge of smoking causing blindness remained similar throughout the later sampling periods in the UK.
Year by year changes in proportions of smokers reporting knowledge of blindness and smoking in Australia are presented in Figure 3. A significant change (χ2 = 14.14, p < 0.001) in proportions reporting knowledge occurred between the Wave 5 sampling period (between October 2006 and February 2007) and Wave 6 (September 2007 to February 2008), coinciding with the introduction of the ‘eye’ warning label. Knowledge among Australian smokers remained high through the subsequent sampling periods between 2007 and 2011 (Waves 7 and 8), peaking at approximately 70 per cent.
Figure 3. The percentage of Australian adult smokers knowing that smoking causes blindness across six waves of the International Tobacco Control Australia Survey (2004 to 2011)
*Indicates significant change in proportion as compared to the preceding wave using χ2 test α = 0.05 significance level
Download figure to PowerPoint
Table 2 presents the findings from the multivariate logistic regression model conducted to examine if demographic and smoking-specific characteristics among adult Australian smokers may be associated with knowledge of smoking causing blindness across all years.
Table 2. Australian smokers' likelihood of reporting knowledge that smoking causes blindness—(International Tobacco Control Australia) 2004–2011
|Before (Waves 1–5)||Base level|| |
|After (Waves 6–8)||1.68 (1.03, 2.76)||0.038a |
| Sex || || |
|Female||Base level|| |
|Male||1.08 (0.78, 1.51)||0.644|
| Age group || || |
|18 to 24||Base level|| |
|25 to 39||0.82 (0.53, 1.28)||0.388|
|40 to 54||0.70 (0.53, 0.93)||0.015a |
|55+||0.43 (0.29, 0.62)||<0.001a |
| Income || || |
|Low income||Base level|| |
|Moderate income||1.25 (0.96, 1.63)||0.095|
|High income||0.86 (0.68, 1.10)||0.224|
|No answer||1.05 (0.72, 1.53)||0.806|
| Education || || |
|Low education||Base level|| |
|Moderate education||1.10 (0.88, 1.37)||0.392|
|High education||1.33 (0.78, 2.26)||0.292|
| Dependence || || |
|Increasing HSI Score||0.99 (0.85, 1.15)||0.89|
Australian smokers in the period post-introduction of the ‘eye’ warning label (September 2007 to May 2011) were 1.68 times as likely (95% CI: 1.03, 2.76, p = 0.038) than Australian smokers in the pre period (June 2004 to February 2007) to report knowledge of the smoking and blindness link (Table 2). In Australia, increasing age was associated with a lower likelihood of knowledge with ages 40 to 54 (odds ratio: 0.70, CI: 0.53, 0.93, p = 0.015) and over 55 years (OR: 0.43, 0.29, 0.62, p < 0.001) significantly less likely than the youngest cohort (18 to 24 years) to report knowledge. Interactions between time of recruitment and all predictors were found to be non-significant (p > 0.05 for all interactions).
Identical regressions conducted separately on each of the other three countries found that there was no significant increase in knowledge between the pre and post periods (p > 0.05 for all); however, in Canada, 25 to 39-year-olds (as compared to the reference of 18 to 24-year-old smokers) reported significantly less knowledge (OR: 0.44, 95% CI: 0.26, 0.76, p = 0.003). In the UK, male smokers (OR: 1.43, 95% CI: 1.07, 1.92), as compared with female smokers, reported higher rates of knowledge. Additionally, in the UK, more highly educated smokers (OR: 2.80, 95% CI: 1.67, 4.42, p < 0.001), as compared with less educated smokers, reported higher rates of knowledge. In the US, lower knowledge was associated with moderate (OR: 0.63, 95% CI: 0.43, 0.91, p = 0.014), highest (OR: 0.50, 95% CI: 0.33, 0.75, p < 0.001) and unknown income levels (OR: 0.47, 95% CI: 0.24, 0.94, p = 0.033), as compared with the lowest income group. Additionally, the most highly educated group was more likely to report knowledge of the smoking-blindness link (OR: 1.61, 95% CI: 1.09, 2.36, p = 0.016) as compared with the least educated American smokers. No other examined factors were found to be associated with knowledge in Canada, UK or the US.
- Top of page
- Grants and Financial Assistance
Prior to the introduction of the ‘eye’ cigarette pack warning label, Australian smokers reported relatively high knowledge of the causal association between smoking and blindness, with approximately half of smokers reporting that smoking can cause blindness. This was significantly higher than knowledge in Canada, UK or the US and is likely due to the success of national social marketing campaigns that included considerable use of a high-impact television advertisement on the topic. The current study finds that the introduction of the Australian ‘eye’ warning label in 2007 is associated with a further increase in the knowledge of adult Australian smokers that smoking causes blindness. In Canada and the US where there were no changes to tobacco product warning labels during the study period, there was no change in knowledge of smoking causing blindness. In the UK, there was a significant increase in knowledge between 2004 and 2005. The reasons for these differences are not clear; however, a national professional education and advocacy program called ‘Smoke gets in your eyes’ was active in the UK between 2004 and 2006 and included efforts to raise the awareness of smoking's impact on ocular health to ophthalmologists and eye-health organisations, patients, the general public and policy-makers. It is important to note that even with these efforts, the vast majority of UK smokers were unaware that smoking causes blindness.
In Australia, knowledge of the causal association about smoking and blindness was not predicted by sex, income, education level or a measure of addictiveness (HSI). The logistic regression model for Australian smokers identified that older Australians, relative to younger, are less likely to report that smoking causes blindness. Analysis revealed that this lack of knowledge was also present prior to the warning labels. Smoking prevention and cessation messaging is important for all ages; however, this finding may help identify a priority for better educating older smokers, especially as the incidence of AMD increases with age.
The year-to-year knowledge of the warning labels also suggests that the annual cycling of the Australian warning labels may impact knowledge. The sampling period October 2008 to July 2009 (Wave 7) of the ITC Four Country Survey best coincides with the time when the Set B labels were in circulation in Australia. In the sampling period between July 2010 and May 2011 (Wave 8), there was a decrease (although not significant) in knowledge, which may be explained by the minimal overlap between survey fieldwork and the ‘eye’ label being in circulation.
Some caution is required in interpreting the overall level of knowledge as the task involved recognition of the problem (that is, agreement to a prompted health risk) rather than unassisted recall. This method is also subject to possible acquiescence bias, but as levels were low in the countries without extensive public education, the difference would appear to be real. Further, respondents were not given any definition of what blindness meant and this may have influenced how participants responded, although this was true across all four countries and cannot be an explanation for the different pre-post pattern of knowledge in Australia compared to the other three countries.
During the study period, non-governmental organisations, including the Macular Degeneration Foundation in Australia engaged in a variety of awareness campaigns that may have helped Australians learn about the causal association between smoking and AMD; however, these campaigns did not primarily focus on this one issue.
Despite Australia's relative success in educating smokers about the association between cigarettes and ocular health, approximately 30 per cent of adult smokers in Australia were either unaware of this important health effect or did not believe the message. The latter possibility is important for ocular health practitioners to consider. When the ‘eye’ warning label was evaluated to inform the Australian government, some participants suggested that the link between smoking and blindness was hard to believe. Some reported having difficulty understanding how smoking affects the eye and how visual loss can result. Thus, there is an opportunity for eye-care practitioners to help educate their patients about the health effects and support national prevention and cessation efforts. In 2010, Optometrists Association Australia adopted a position statement on tobacco that states that optometrists have a responsibility to advise their patients of the risks of tobacco smoking and to encourage smokers to stop smoking.
There have been numerous calls for the optometric community to be engaged in tobacco use cessation and prevention work[30, 31] and evidence exists that optometrists are open to being more involved.[32, 33] Integrating the Australian optometric community more meaningfully into the national cessation referral system may be a valuable way to improve national quit rates and address disbelief about smoking causing visual loss. Other countries should take heed and increase their efforts to educate smokers about the link between their smoking and the risk of losing sight.
Grants and Financial Assistance
- Top of page
- Grants and Financial Assistance
Funding support for the ITC Four Country Survey was provided by the National Health and Medical Research Council of Australia (265903 and 450110), Commonwealth Department of Health and Ageing, Canadian Institutes for Health Research (57897, 79551, and 115016), U.S. National Cancer Institute (P50 CA111236 and RO1 CA100362), Robert Wood Johnson Foundation (045734), Cancer Research UK (C312/A3726, C312/A6465, C312/A11039 and C312/A11943) and the Ontario Institute for Cancer Research (Senior Investigator Award). Additional support was provided from the Canadian Cancer Society through the Propel Centre for Population Health Impact and a Canadian Institutes of Health Research New Investigator Award and the Canadian Cancer Society Research Institute Junior Investigator Award (Hammond).