Public support for tobacco control policy extensions in Western Australia: a cross-sectional study.

OBJECTIVES: Policy makers seeking to introduce new tobacco control measures need to anticipate community support to assist them in planning appropriate implementation strategies. This study assessed community support for plain packaging and smoking bans in outdoor locations in Australia.
DESIGN: Analytical cross-sectional survey. SETTING AND PARTICIPANTS: 2005 Western Australian adults participated in a computer-assisted telephone interview. Random household telephone numbers were used to obtain a representative sample. OUTCOME MEASURES: Support for plain packaging of cigarettes and smoking bans at outdoor venues by demographic characteristics.
RESULTS: Around half of the survey respondents supported plain packaging and almost a further quarter reported being neutral on the issue. Only one in three smokers disagreed with the introduction of a plain packaging policy. A majority of respondents supported smoking bans at five of the six nominated venues, with support being strongest among those with children under the age of 15 years. The venues with the highest levels of support were those where smoke-free policies had already been voluntarily introduced by the venue managers, where children were most likely to be in attendance, and that were more limited in size.
CONCLUSIONS: The study results demonstrate community support for new tobacco control policies. This evidence can be used by public policy makers in their deliberations relating to the introduction of more extensive tobacco control regulations.

Introduction

Tobacco control policies to date have encompassed various elements including cessation assistance, taxation, restrictions on tobacco advertising and removal of point of sale displays, tobacco packaging regulations and designation of non-smoking areas.1 Most tobacco control policies have considerable public support across varying countries and among both smokers and non-smokers.2–12 As the evidence base relating to the harms associated with secondhand smoke grows and tobacco consumption becomes increasingly de-normalised in many countries, some governments are stepping up their efforts to further discourage smoking at a population level.13 14 Two areas of possible tobacco control policy extension include plain packaging of cigarettes and smoking bans in outdoor locations.1 2 14

In line with work demonstrating that the visual characteristics of cigarette packages convey important information to consumers and influence their perceptions of the healthiness and desirability of the product,15 recent research indicates that plain packaging may reduce demand for cigarettes. In Australia, Wakefield and colleagues16 17 demonstrated that as branding symbolism is reduced on packages, associations of the types of people who would use the product and assessments of the appeal of the product become progressively more negative. Similarly, research by Hoek et al18 with young adult smokers in New Zealand indicated that the brand information displayed on cigarette packages can be important for identity construction and communication for this age group, and hence that the minimisation of this information and the inclusion of more prominent health warnings can reduce the attraction of the product. Attitudinal research conducted in the USA,19 Canada,20 the UK15 21 and France22 has reached the same general conclusions.

In a world first, legislation has been passed in Australia that requires plain packaging to be phased in on all cigarette packages during 2012.23 24 There has been considerable support for plain packaging among key stakeholders, including public health advocates and the three primary political parties.1 25 However, anecdotal evidence suggests that the issue is more contentious among the general public, an outcome that is at least partly attributed to concerted efforts by the tobacco industry to incite public opposition to the change.25 Research is therefore necessary to gauge levels of support for plain packaging to assist governments in planning appropriate implementation strategies.

Restrictions on smoking in public places have been found to be effective in terms of achieving compliance among smokers and reducing exposure to secondhand smoke.4 They can also contribute to the prevention of smoking uptake among children and young people by reshaping the perceived social acceptability of smoking.26 The ability of smoke-free policies to normalise non-smoking is evident in the dramatic increases in support for such policies that have occurred in numerous countries post-implementation.4 27–29 Of note is that many of these policies have been introduced on the basis of research evidence despite only minority support pre-implementation.

While numerous countries have implemented smoking bans in workplaces, restaurants and bars,27 30 and in some cases the outdoor areas adjacent to these locations,14 other outdoor locations remain mostly unregulated. To date, there appears to have been little research on the extent of support for bans in outdoor locations characterised by open spaces that are not directly adjacent to buildings. In their review of the few studies that have explored this issue, Thomson and colleagues31 noted that there are typically high levels of community support but that further research is needed to more comprehensively assess the extent of support for such bans and how this may vary in different contexts.31

While high levels of community support are not a prerequisite for successful policy implementation due to the normalisation outcomes noted above, they are likely to be of value in facilitating policies through the relevant legislative processes and minimising the costs associated with community consultation and education. Evidence relating to levels of support for new tobacco control policies is useful for informing governments of (1) the extent to which legislative changes are aligned with community attitudes and (2) any variations in attitudes among different stakeholder groups. This information can assist in ensuring that new policies are effectively communicated to enhance acceptance and compliance.2 32 Awareness of public attitudes is particularly important in the context of sophisticated lobbying and public relations activities by the tobacco industry33 and the potential for media coverage to influence support for tobacco control policies.34 35

To this end, the present study investigated public support for plain packaging of cigarettes and smoking bans at a range of outdoor venues. Australia, the context of the study, is acknowledged to be one of the world leaders in tobacco control interventions.6 7 30 Attitudes to smoke-free policies have become more positive over the last decade,14 and the proportion of the adult population classified as current daily smokers has decreased to 19% nationally and 12% in Western Australia.36 37 These outcomes have been attributed to ongoing public education campaigns and other tobacco control strategies implemented since the 1970s.14 38 39 The purpose of the study was to investigate whether the population is amenable to further policy extensions.

The study was funded by the Western Australian Health Promotion Foundation. Clearance was obtained from the UWA Human Ethics Committee and no competing interests were declared.

Method

Sample and survey design

Data for this study were collected via a household telephone survey in November 2010. Respondents were residents of Western Australia aged 18–69 years. Telephone numbers were randomly selected from an electronic household telephone directory. Although the directory included only landlines (ie, mobile phone numbers are not listed), the rate of landline ownership in Australia has remained high over the last decade at around 88%.40 The adult in the house with the next birthday was invited to participate in the survey. Up to 10 call backs were made to each household to maximise the response rate, resulting in a response rate of 60% among eligible households. The final sample comprised 2005 respondents, 66% of whom resided in the Perth Metropolitan area and the remainder in country areas. To adjust for any differences from the general population, the sample was weighted to the age and location distribution of the Western Australian population aged 16–69 years according to 2006 census data.41

The survey instrument comprised a range of questions relating to lifestyle and attitudes to smoking. One item asked respondents whether they agreed, disagreed or had no feelings about the introduction of plain packaging on cigarettes. The Australian government had announced its intention to introduce plain packaging in April 2010,25 and it was likely that many respondents were aware of this at the time of the survey. Other survey items asked respondents whether they supported, did not support or had no feelings either way about smoke-free policies in six outdoor venues: the Perth Zoo, the South Perth Foreshore Family Zone (a riverfront venue for music festivals), the Perth Royal Show (an annual agricultural event), Adventure World (a theme park), Kings Park (a national park located adjacent to the city centre) and Rottnest Island (a holiday resort island located 20 km from the mainland). The purposefully selected venues represented a wide range of locations, including some that had elected to be smoke freei and others that had no smoking policies at the time of data collection. Some of the venues were family oriented, while others attracted the general population. Some were permanent fixtures, while others were annual or seasonal events. Almost all the venues, with the exception of the Royal Show, had an estimated 90% or more open space within the venue boundaries. Table 1 provides a description of each of the venues included in the study.

Table 1

Venues and venue attributes

Venue	Venue attributes
	Land size	Family or general population focus	Smoke-free policy status	Frequency of event
Perth Zoo	41 acres	Family	Smoking only permitted in dedicated smoking areas	Continual
South Perth Foreshore Family Zone	30 acres	Family	Total smoking ban	Annual
Perth Royal Show	75 acres	Family	Total smoking ban	Annual
Adventure World theme park	660 acres	Family	Total smoking ban	Summer only
Kings Park	1003 acres	General	No policy	Continual
Rottnest Island	31 179 acres	General	No policy	Continual

Data analysis comprised descriptive statistics to ascertain attitudes towards the two proposed tobacco control policies. The χ2 statistic was used to determine significant association between policy support and participant characteristics (age, gender, smoking status and having a child under 15 years of age). Multinomial logistic regression models were used to generate ORs for agreement/no agreement with the proposed policies, adjusting for respondents who ‘had no feelings either way’ and the participant characteristics explored in the descriptive analyses. All statistical analyses were performed using SPSS V.19.0.

Results

The sample comprised 1016 men (50.7%) and 989 women (49.3%), with 29.3% (n=588) aged 16–29 years, 42.0% (n=842) aged 30–49 years and 28.7% (n=575) aged 50–69 years. Almost two-thirds of respondents (63.5%; n=1273) had no children under 15 years of age, and around half (53.5%; n=1073) had completed post-secondary education. The sample comprised 12.3% (n=344) current smokers, 22.6% (n=473) ex-smokers and 59% (n=1189) who had never smoked. A significantly higher proportion of men than women were current smokers (21.9% vs 12.3%; p<0.00).

Table 2 shows the overall and subgroup levels of support for a plain packaging policy. Around half the total sample were in agreement, a quarter disagreed and a further quarter reported that they had no feelings either way. Agreement levels were lower among smokers, with approximately a third in agreement, a quarter neutral and the remainder expressing disagreement. Non-smokers (including ex-smokers) were 2.7 times (95% CI 2.0 to 3.6) more likely than smokers to agree, while smokers were 2.6 times (95% CI 1.9 to 3.6) more likely to have no feelings either way. Men were more likely than women to agree with plain packaging (53.3% vs 47.8%; OR 1.4, 95% CI 1.1 to 1.7) and less likely to have no feelings either way (21.3% vs 24.4%; OR 0.9, 95% CI 0.7 to 1.2). Compared with respondents aged 16–29 years (42.0%), respondents aged 50–69 years (55.1%) were 2.0 (95% CI 1.5 to 2.7) times more likely to agree with the introduction of plain packaging.

Table 2

Support for plain packaging

	Total (n=2005)	Gender*		Smoking status**			Children**		Age**
		Male (n=1016)	Female (n=989)	Non-smoker (n=1189)	Ex-smoker (n=472)	Smoker (n=343)	No child under 15 (n=1273)	Child under 15 (n=732)	16–29 years (n=575)	35–49 years (n=842)	50–69 years (n=588)
	%	%	%	%	%	%	%	%	%	%	%
Agreed	51	53	48	55	52	34	49	53	55	53	42
Disagree	27	25	28	25	21	38	29	22	23	24	35
No feelings either way	23	21	24	20	26	29	22	25	22	23	23

Significant associations at the p<0.05 level.

Significant associations at the p<0.01 level.

As can be seen in table 3, support for smoking bans was strong for four of the six specified outdoor locations. Those who had never smoked and those with children under 15 years of age demonstrated the highest levels of support for smoking bans at all six venues. Of note is that support was strong among smokers for some venues, especially the South Perth Foreshore and the Zoo.

Table 3

Support for outdoor smoking bans by venue and respondent characteristics

Venue	Total (n=2005)	Male (n=1016)	Female (n=989)	Non-smoker (n=1189)	Ex-smoker (n=472)	Smoker (n=343)	No child under 15 (n=1273)	Child under 15 (n=732)
	%	%	%	%	%	%	%	%
Perth Zoo*†‡	85	81	89	90	84	68	83	88
South Perth Foreshore Family Zone*†‡	82	80	84	85	82	72	80	85
Perth Royal Show†	80	77	83	96	78	60	77	84
Adventure World theme park*†‡	77	74	81	83	77	58	75	81
Kings Park†	52	52	54	62	49	26	52	53
Rottnest Island†	43	42	44	51	40	19	42	45

Significant associations at the p<0.01 level for each venue are represented as for gender.

Significant associations at the p<0.01 level for each venue are represented as for smoking status.

Significant associations at the p<0.01 level for each venue are represented as for having a child under 15.

Table 4 provides the ORs for the variables of gender, age, parental status and smoking status. Respondents' smoking status was the strongest indicator of likely support, with significant and large differences in the ORs of ex-smokers and non-smokers compared with smokers for each of the venues. Independent of smoking status, women were more likely than men to support the Perth Zoo, the South Perth Foreshore and the Perth Royal Show becoming smoke-free venues. Independent of smoking status, gender had little impact on support for smoke-free venues, with the exception of women who were 1.3 times more likely than men to support the Zoo being a non-smoking venue. Similarly, age was observed to have little independent effect on support for smoke-free venues, with the exceptions of Adventure World, where a smoking ban was supported significantly more by respondents aged 30+ years, and Kings Park, where respondents aged 30–49 years were less likely to support a smoking ban.

Table 4

Support for smoking bans at nominated outdoor venues

	Zoo	South Perth	Royal Show	Adventure world	Kings Park	Rottnest
	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)
Male	–	–	–	–	–	–
Female	1.5 (1.1 to 2.1)	1.7 (1.1 to 2.5)	1.5 (1.1 to 2.0)	NS	NS	NS
16–29	–	–	–	–	–	–
30–49	NS	NS	NS	1.6 (1.1 to 2.3)	0.7 (0.6 to 1.0)	NS
50–69	NS	NS	NS	2.2 (1.5 to 3.3)	NS	NS
No child under 15	–	–	–	–	–	–
Child under 15	NS	NS	1.9 (1.4 to 2.8)	1.4 (1.0 to 1.9)	NS	1.3 (1.0 to 1.7)
Smoker	–	–	–	–	–	–
Ex-smoker	3.1 (2.0 to 4.4)	NS	3.3 (2.3 to 4.8)	2.4 (1.7 to 3.5)	3.0 (2.2 to 4.1)	3.2 (2.3 to 4.6)
Never smoked	7.2 (5.0 to 10.3)	2.9 (1.8 to 4.6)	7.1 (5.1 to 9.9)	6.0 (4.3 to 8.4)	5.6 (4.2 to 7.4)	6.2 (4.6 to 8.4)

Respondents with children under 15 years of age were more likely than other respondents to support smoking bans at the South Perth Foreshore, the Royal Show and Adventure World. Women with children under 15 years of age were 1.6 (95% CI 1.08 to 2.49) times more likely than other women to support smoking bans at the Zoo. No significant independent relationship was observed for parental status and support for the introduction of bans at Kings Park, the Zoo or the South Perth Foreshore.

Along with respondent characteristics, it is likely that venue attributes influenced support levels. Respondents were more supportive of smoke-free policies for venues that already had smoking restrictions in place, were smaller in size and had a family focus. In general, as venue size increased and the target population attending the venues became more representative of the general public, support weakened.

Discussion

The present study investigated public attitudes to potential tobacco control policy extensions, namely plain packaging requirements and smoking bans at outdoor venues. In terms of community support for plain packaging, approximately half of the survey respondents supported the initiative and almost a quarter were neutral. This suggests that there is unlikely to be a substantial public backlash when plain packaging is introduced in Australia. Of note is that only one in three current smokers, who constitute 12% of the Western Australian adult population, disagreed with the policy. In the light of a primary aim of plain packaging policy being to discourage new adopters,14 42 this may reflect high levels of regret for smoking initiation among current smokers.43 44

As governments increasingly legislate against smoking in workplaces and other indoor locations, extension to outdoor contexts is the next frontier. Smoking bans in outdoor settings may have multiple benefits in terms of reducing non-smokers' exposure to secondhand smoke, encouraging smokers to reduce their intake or quit and avoiding instances of adults modelling and normalising tobacco consumption to children.45 However, there is a need to better understand community support for such policy extensions to optimise implementation and compliance. Reflecting the findings of limited previous work in this area,31 the results of the present study show majority support for smoking bans at five of the six nominated venues. The one exception was an island resort that is large in acreage and patronised by both local residents and tourists. The tendency for support to be highest among venues that already have voluntary smoking bans in place is aligned with previous research suggesting that support for bans increases post-implementation.27 28 30

Past research has found women to be generally more supportive of smoking policies than men.6 7 32 In the present study, women tended to be more supportive of smoking bans in outdoor venues, but men were more supportive of plain packaging. The reasons for women's lower support for plain packaging are not clear and would require further research to understand whether this outcome relates to weaker beliefs about the potential effectiveness of this approach or stronger involvement with brands per se.46 More advanced age has been associated with greater support for smoking policies elsewhere,32 and this was found to be the case for plain packaging and smoking bans at some of the nominated venues but not all. It thus appears that demographic attributes may have varying relationships to support levels in different national contexts and that country-specific research needs to be undertaken to assess likely reactions to policy changes among different population segments.

Other areas of possible future research include (1) investigation of public support for plain packaging legislation once it has been fully implemented, (2) analysis of venue management and community support for voluntary versus mandatory smoke-free policies in outdoor areas and (3) assessment of the impact of media coverage of proposed tobacco control policies on levels of community support. In addition, further research analysing associations between smoke-free policies and children's normative beliefs about smoking (such as their estimates of peer or community prevalence and views on social acceptability) would be a valuable contribution to youth smoking prevention research.

Despite some variations in support for plain packaging and outdoor smoking bans by demographic characteristics, the overall levels of support for these policies suggest that it would not be necessary to differentially accommodate age and gender in communications strategies designed to educate the community about their introduction and implementation. However, the large variations in support by smoking status indicate that smokers are a specific communications audience that may require special consideration. For example, information relating to the types of outdoor venues affected by any new smoke-free policy would need to be broadly disseminated to prevent smokers from arriving at these venues without being able to prepare for abstinence during attendance.

In conclusion, the results of the present study demonstrate community support for more stringent tobacco control policies in Australia. This evidence may be of use to public policy makers in their deliberations relating to future extensions of existing regulations. Further research is needed in other countries to assess the extent to which the results reflect prevailing community views in other locations.