Ethics approval

This study was approved by the Human Research Ethics Committee at the University of Sydney [No. 2017/202]. Before commencing the study, participants were asked to read a Participant Information Sheet and an informed consent statement. Participants were required to indicate their consent by clicking ‘Yes’ on an electronic consent form before the survey could begin. Participants were informed about their right to exit the survey at any time, and all participant data were analysed and reported anonymously.

Participant recruitment

An online market research survey was administered to a cross-sectional sample of Australian residents aged ≥18 years in May 2017. Market research company Online Research Unit (ORU) was contracted to recruit participants and distribute the survey. Participants were recruited using both online (email, internet) and offline (telephone, print and postal) methods to ensure broad demographic representation, and to limit recruitment of professional survey respondents. ORU used stratified sampling to ensure respondents were nationally representative for age, gender and location. The survey was run over 12 days until the minimum 2000 responses were received. Participants were rewarded points worth a nominal value ($2.00) upon completion of the survey.

Survey design

The survey, hosted through secure web application Redcap, measured respondents’ perception of the obesity problem in Australia, the responsibility of governments, and level of agreement on a 5-point Likert scale (strongly disagree to strongly agree) with proposed government regulations (Additional file 1). Proposed regulatory approaches included restrictions on food and beverage advertising, prohibiting food and beverage sponsorship of children’s sport and food and beverage taxation. Participants were also asked to respond to seven demographic questions including gender, age, education level, postcode, weight, height and parental status. The survey questions were developed in consultation with an advisory group of public health experts experienced in market research surveys.

A pilot version of the survey was conducted with 166 respondents and the data analysed to measure the scale reliability and validity of the survey tool. Principal components analysis was carried out using IBM Windows SPSS version 22.0 [43] on the questions measuring support for proposed government regulations (question 3–13). Two components recorded initial eigenvalues greater than 1.0, explaining 76.1% of the variance: (1) support for advertising and marketing restrictions; and (2) support for food and beverage taxation. Internal consistency within each component was evaluated using Cronbach’s α. Both components had high internal consistency, with Cronbach’s α of 0.92 and 0.95 for components 1 and 2 respectively. The effect of removing individual questions on the Cronbach’s α was calculated and no redundant questions were identified. Only one error was detected during data cleaning (question 15 was missing an age-range category) and this question was corrected before the official survey launch. Detailed results of the pilot survey are available from the corresponding author on request.

Data analysis

Participant characteristics

A total of 2135 participants accessed the survey, with 2011 complete and 124 partial responses. Table 1 shows the demographic characteristics of participants. The gender, age, location, BMI and parental status of participants were reflective of the Australian population, based on ABS 2011 Census Data. Survey participants were more highly educated than the general population, with 72.2% completing a tertiary education qualification (University degree, Technical and Further Education [TAFE], trade or diploma) compared to only 57.0% of the Australian population. Data were weighted for education level in analysis.

Perception of the overweight and obesity problem in Australia

The majority of participants (92.5%) considered overweight and obesity to be a somewhat or very serious problem in Australia, with 5.1% perceiving it as not too serious and 1.0% perceiving it as not at all serious (Table 2). Males were significantly less likely than females to consider overweight and obesity to be a serious issue [Odds ratio (OR) 0.46, 95% 0.24–0.85 p = 0.01] (Additional file 2). Parental status also influenced opinions, as participants without children were significantly less likely than participants with children to agree that overweight and obesity is a serious problem in Australia [OR 0.43, 95% CI 0.23–0.82, p = 0.01]. Very few participants felt there was no need for government regulation to address overweight and obesity (9.4%), with the remaining participants believing there should be a little (20.5%), some (43.5%) or a great deal (22.1%) of regulation. University educated participants were significantly more likely than those who did not complete school to feel more government regulation was needed [OR 1.68, 95% CI 1.24–2.28, p = 0.001].

Public support for regulation of food and beverage advertising

Food and beverage advertising regulations were the best supported of the policy options proposed in the questionnaire (Table 2). Support was highest for restrictions on unhealthy food advertising directed at children on television (78.9%) and the internet (75.8%). Females were consistently more likely than males to agree that the government should regulate unhealthy food and beverage advertising, irrespective of the medium (p < 0.05) (Additional file 2). Support for advertising restrictions decreased as SEIFA score increased, with those in the highest SES tertile (i.e. from areas of highest socioeconomic advantage) significantly less likely than those in the lowest SES tertile to support government regulation of unhealthy food and beverage advertising on television (OR 0.50, 95% CI 0.33–0.77, p = 0.001], the internet [OR 0.53, 95% CI 0.36–0.79, p = 0.002], in public spaces [OR 0.52, 95% CI 0.36–0.76, p = 0.001] and at sporting events [OR 0.50, 95% CI 0.34–0.71, p < 0.001]. Respondents with a tertiary education qualification were also more likely to support proposed advertising restrictions (p < 0.05).

Public support for regulation of children’s sport sponsorships

Fifty-nine percent of participants agreed with prohibiting fast food company sponsorship of children’s sport, and 63.4% supported prohibitions on SSB company sponsorship (Table 2). As seen in Additional file 2, education level was a consistent predictor of support, with tertiary educated respondents significantly more likely than those who did not complete high school to agree with government regulation of children’s sport sponsorships (p < 0.001). Males were also significantly less likely than females to support prohibitions on fast food company sponsorship [OR 0.65, 95% CI 0.51–0.83, p < 0.001] and SSB company sponsorship of children’s sport [OR 0.68, 95% CI 0.53–0.87, p = 0.002].

Public support for food and beverage taxation

There was least support among survey respondents for a tax on unhealthy foods (48.9%) and SSBs (54.5%); however, there was a slight increase in regulatory support if the revenue generated from the tax is used for health purposes (Table 2). Age was a predictor of attitudes towards regulation, with younger respondents (18–24 years) more supportive of a hypothecated tax (p < 0.05) (Additional file 2). Respondents who completed a university degree were significantly more likely than those who did not complete high school to support taxation of unhealthy foods [OR 2.27, 95% CI 1.63–3.15, p < 0.001] and SSBs [OR 2.74, 95% CI 1.95–3.84, p < 0.001]. Respondents without children were significantly less likely to support government taxation of unhealthy foods [OR 0.78, 95% CI 0.62–0.99, p = 0.04], although this relationship was not observed for SSB taxation.

Regulation of food and beverage advertising

The results of this survey demonstrate that there is substantial public support for regulations to reduce exposure to unhealthy food and beverage advertising in Australia, in particular broadcast advertising. Nearly four in five respondents believed that the government should restrict unhealthy food and beverage advertising during times when children are watching television, and three in four supported restrictions on unhealthy food and beverage advertising to children on the internet. Earlier Australian studies reported slightly higher levels of support for restrictions on unhealthy food advertising directed at children (83%), although this was likely due to sampling differences as the studies predominantly surveyed females [32, 33]. There is a growing body of literature highlighting the direct causal link between advertising exposure and the taste preferences, purchasing preferences, and consumption of unhealthy food by children [6, 45–48], and parental concerns are rising about the high volume of unhealthy food advertising directed to children on television [49, 50]. As community awareness about the negative influence of food advertising builds, it may be expected that support for regulation will continue to rise.

The results of this survey also indicate strong support (> 70% agreement) for restrictions on unhealthy food advertising at sporting events and in public areas such as transit stops. Advertisements in public settings have the potential to significantly influence the dietary behaviours of both children and adults, given that these areas are highly frequented and the nature of the viewing is non-discretionary. Recent Australian research demonstrates that over 80% of food and beverage advertising at train stations and in the areas surrounding schools are for unhealthy products [51–53], suggesting that government regulation of these settings is also warranted.

Regulation of children’s sport sponsorship

Nearly two out of three respondents supported regulations banning fast food and SSB companies from sponsoring children’s sport (63.4% support overall); a similar result to previous Australian surveys [39, 42]. The community appreciates that sporting clubs require financial assistance, and that sponsorship can assist in subsidising cost of participation and reducing equipment and resource costs [39]. A public survey conducted by the Victorian Health Promotion Foundation in 2009 (n = 1500) found that 53% of respondents opposed junk food sponsorship of community sports clubs, but support for removing sponsorship increased to 81% if the government were to replace lost revenue [54]. The results may also reflect the parental perception that children are less influenced by sponsorship of amateur sport compared with elite sport sponsorship [39], even though studies have shown that children have high brand recall and positive attitudes towards food and beverage sponsorship companies [55]. Parents were as likely as non-parents to agree with the proposed regulations, an unexpected but promising finding suggesting broad community concern for the health of children.

Taxation of unhealthy foods and beverages

In line with previous local and international research, taxes on unhealthy food and SSBs were less well-supported than advertising restrictions [56–58], although support was slightly higher if the revenue was used for public health purposes. There is continuing public debate between health experts, policymakers, and the food and beverage industry about the merits of a tax on unhealthy foods and beverages. Key anti-tax messages that have been circulated by the food and beverage industry include the potential loss of revenue and employment, the regressive nature of the tax for lower SES groups, and the ineffectiveness of taxes in improving unhealthy dietary behaviours [59]. In addition, taxes are commonly framed as “nanny-state” interventions that restrict individual freedom of choice [60]. A recent study which assessed public opinion in the US towards arguments used in tax debates showed that the public more strongly agreed with anti-tax messages compared with pro-tax messages [61], indicating a need to re-frame and strengthen the pro-tax messages that are disseminated.

Despite taxation being the least favoured policy option, there was still considerable public support for a tax on SSBs, particularly if the money raised is used for health-related purposes (57.2%). Consumption of SSBs in Australia is high [62], and there is an established association between rising consumption of SSBs and rising rates of obesity, type 2 diabetes and dental caries [63, 64]. What constitutes a sugar-sweetened beverage is easy to define, whereas a tax on unhealthy foods is more ambiguous, and could potentially have unintended health effects due to patterns of food substitution in response to the tax [65].

Determinants of public support for regulation

The results of this survey highlight the influence of demographic factors, in particular gender and education level, on support for government regulation. Despite no significant between-group difference in the perceived severity of the obesity problem, respondents with a tertiary education were significantly more supportive of all proposed regulations, in line with existing literature [49, 57, 66, 67]. It is hypothesised that this is due to educated respondents having a greater awareness and understanding of the environmental influences on health. While it was expected that higher SES groups would also be more supportive of government regulation due to the known correlations between education level and SES, this was not observed, with lower SES respondents more supportive of advertising restrictions.

Females had a tendency to be more supportive of regulation of the food environment, whether because they are more health conscious or because they are more likely to be the household’s primary grocery buyer, with greater appreciation of the challenges of purchasing healthy food. Younger respondents aged 18–24 years were also significantly more likely to support taxation of SSBs and unhealthy foods compared with older adults, particularly if the money was to be used for health-related purposes. Our results are consistent with those of Rivard et al., who observed greater support for an SSB tax among younger respondents [66]. This age-group is the largest consumer of SSBs in Australia [62], and would be expected to benefit most from a tax.

Limitations

Unlike other studies which examined determinants of support for obesity prevention policies, this survey did not explore the public’s beliefs about the causes of overweight and obesity. Previous literature has suggested that those who perceive environmental factors to be the biggest driver of overweight and obesity are significantly more supportive of government regulation compared to those who attribute it to individual choice [28]. Further, this survey did not explore why certain regulations were not supported (e.g. whether viewed as intrusive, or ineffective in changing behaviours). The grouping of participants into BMI categories was based on self-reported height and weight. Given the tendency for people to underreport these values, we are unable to accurately assess whether BMI is associated with support for regulation.

Implications for future action

In Australia, the federal government’s constitutional powers enable it to regulate broadcast (including television and radio) and digital advertising, although state governments retain regulatory power over outdoor advertising – a prime setting for unhealthy food and beverage advertising [51]. Under the co-regulatory framework governing broadcasting regulation, the Australian Communication and Media Authority (ACMA) is legally authorised to register and monitor codes of practice developed in conjunction with industry groups, a number of which contain controls on advertising [68]. One model for regulating broadcast advertising would therefore be to introduce a code imposing appropriate constraints on unhealthy food and beverage advertising that is registered with the ACMA, becoming part of the licence conditions for broadcasting licensees. However, due to the variety of advertising formats and platforms available, regulation of advertising on the internet and on mobile communications is likely to require controls to be extended beyond broadcast licensees. For example, federal legislation already prohibits the publication of tobacco advertisements in Australia, including advertisements that are accessed or disseminated through the internet, computers, mobile devices and other electronic devices [69].

Currently in Australia there is no policy or regulation that limits or restricts the types of companies allowed to sponsor children’s sport. Before prohibitions on food company sponsorship can be acceptably introduced by the government, there needs to be consideration of how children’s sports clubs will continue to be funded. Pettigrew et al. measured what sponsors the public would consider acceptable at children’s sport, identifying health agencies, sporting goods companies, banks/finance companies and bottled water companies as suitable options [42]. Alternatively, Kelly et al. proposed the establishment of an intermediary Sport Sponsorship Fund that could manage the collection and distribution of corporate funding for sports clubs; a strategy that was well supported by parents [70]. State government health agencies could also work as funding bodies for sporting clubs: the ACT, South Australia, Victoria and Western Australia adopted this approach following the introduction of the Federal Tobacco Advertising Prohibition Act in 1992 [71].

More than half of respondents in our survey supported a tax on sugar-sweetened beverages, and nearly half supported a tax on unhealthy foods. There are a number of tax models for consideration by federal government; specific excise taxes being the most widely accepted and used [19]. An excise tax on SSBs, if implemented at a tiered rate as in England, has the potential to reduce sugar intake, generate government revenue, educate consumers and drive the reformulation of beverages [72]. Although one in four survey respondents disagreed with introducing a tax on unhealthy foods and beverages, literature has shown that support for public health policies such as smoking bans [73], tobacco taxation [26] and mandatory seat-belt legislation [74] can increase post-policy implementation. A recent study measuring public acceptance of the SSB tax in France (n = 1996) also reported an increase in support post-policy implementation [57]. Public acceptance of taxes was shown to increase if the money raised is used for health purposes, so marketing taxes as a revenue-raising tool rather than a preventative health initiative is likely to build both public and political support. Additionally, more advocacy work is needed to educate the public on the positive health, economic and social impacts of SSB taxes, as a significant number of survey respondents appeared to have no opinion on the issue (17.4% neither agree nor disagree). Advocacy and education needs to be carefully targeted to the right demographic groups, in particular males and those without a tertiary education.