414 Australian and New Zealand Journal of Public Health 2014 vol. 38 no. 5© 2014 Public Health Association of Australia

Diets inclusive of excessive quantities of energy-dense, nutrient-poor foods and sugar-sweetened beverages are

a key contributor to weight gain.1 In 2011–12, 62.8% of adult Australians were classified as overweight or obese2 and these rates are reported to be higher among women in more socioeconomically disadvantaged areas.3

Environmental factors such as the neighbourhood food environment are increasingly recognised as potential contributors to obesity rates.4,5 These may be particularly relevant when considering explanations for health disparities between different socioeconomic areas.6

Advertising influences food preferences, purchasing and consumption.7-9 Internationally, there are various forms of food advertising and studies have reported that outdoor advertising usually promotes unhealthy foods.10,11 This being the case, advertising is likely to have played a role in the current obesity epidemic.12-14 Furthermore, targeted advertising of unhealthy foods may entrench and even increase existing socioeconomic inequalities in the prevalence of obesity.15

Internationally, there is limited literature on the prevalence of outdoor food advertising, including the use of billboards and wall-posters to promote food/beverages within the community.10,11,16-22 The placement of outdoor food advertisements in prominent areas such as public transit stops means viewing is not discretionary when compared to advertising on television or other media. Most previous studies of outdoor food

advertising have concentrated on the density of outdoor food advertisements around schools,10,11,16,18 while some have focused on area-level socioeconomic disadvantage or ethnic disparities.11,17,18,21

Australian studies from Perth and Sydney previously investigated advertising at public transit stops.22,23 Kelly et al. assessed food and beverage advertisements as part of their investigation into the products available in vending machines at train stops in Sydney, finding that unhealthy snacks were commonly advertised, although water was the most common beverage item advertised.22 Pierce et al. focused on bus stops

in Perth and reported that 172 (23.1%) of the audited stops had advertisements for alcohol products.23 Neither of these studies assessed all forms of public transport stops and they did not assess differences relative to area-level socioeconomic disadvantage.

In July 2011, a House of Representatives committee tabled a report on the regulation of billboards and outdoor food advertising with key recommendations aimed to ensure advertisements met community standards.24 However, there remains little published data within Australia as to the current content of these advertisements. To better understand the distribution and type of outdoor food

Socioeconomic differences in outdoor food advertising at public transit stops across Melbourne suburbsPhilippa J. Settle,1 Adrian J. Cameron,2 Lukar E. Thornton1

1. Centre for Physical Activity and Nutrition Research, School of Exercise and Nutrition Sciences, Deakin University, Victoria2. World Health Organization Collaborating Centre for Obesity Prevention, Deakin Population Health, VictoriaCorrespondence to: Dr Lukar Thornton, School of Exercise and Nutrition Sciences, Deakin University, 221 Burwood Highway, Burwood, Victoria, 3125;

e-mail: lukar.thornton@deakin.edu.auSubmitted: January 2014; Revision requested: March 2014; Accepted: April 2014The authors have stated they have no conflict of interest.

Aust NZ J Public Health. 2014; 38:414-8; doi: 10.1111/1753-6405.12257

Abstract

Objective: To assess and compare the number and type of outdoor food advertisements at public transit stops within suburbs of varying levels of socioeconomic disadvantage.

Method: An observational audit tool was developed and implemented to assess the number and type of food advertisements at public transit stops within Melbourne, Victoria. A total of 20 Melbourne neighbourhoods (suburbs) from across the least and the most socioeconomically disadvantaged areas were selected. All public transit stops, including train stations and bus and tram stops with a shelter were audited. 

Results: A similar proportion of transit stops in the least and most-disadvantaged suburbs displayed food advertisements (total n=203). However, some differences in the type of advertisements across suburbs were noted with advertisements for fast food restaurants, flavoured milk and fruit juice more common in the most-disadvantaged neighbourhoods (all p<0.05) and advertisements for diet soft drink, tea, coffee and convenience stores more common in the least-disadvantaged neighbourhoods (all p<0.05).

Conclusion: This exploration of outdoor food advertising at Melbourne transit stops found 30% displayed food advertisements, with those in more disadvantaged suburbs more frequently promoting chain-brand fast food and less frequently promoting diet varieties of soft drinks. These findings may help raise awareness of unhealthy environmental exposures.

Key words: food advertising, area-level disadvantage, public transport

SOCIOECONOMIC STATUS AND HEALTH

2014 vol. 38 no. 5 Australian and New Zealand Journal of Public Health 415© 2014 Public Health Association of Australia

advertisements, the current study examined the prevalence of outdoor food advertising at tram, bus and train public transit stops across the least and most socioeconomically disadvantaged areas of Melbourne, Australia.

Methods

Sampling The sampling area for this study was the region within a 20 km radius of the Deakin University Burwood campus. There were 253 suburbs identified in this region. The Australian Bureau of Statistics (ABS) Socioeconomic Index for Areas (SEIFA) Index of Relative Socioeconomic Disadvantage (IRSD)25 score and state decile ranking were sourced for each of these suburbs. A total of 38 suburbs within the region were considered to be in the most-disadvantaged quintile while 96 were from the least-disadvantaged quintile.

A random selection of 10 suburbs from both the most-disadvantaged and least disadvantages quintiles was made, with the absence of a train station within the suburb being the only exclusion criteria.

Audit toolThe auditing tool used for this study was designed by the authors and guided by measures recorded in other studies.10,11,16-22 The tool was refined for ease-of-use following pilot audits. The products in the advertisements were classified into nine categories including cold beverages, hot beverages, snack foods, fast food, breakfast cereal, fruits and vegetables, dairy, other food products and food stores. Within these, sub-categories related to the specific items were listed (see the supplementary file available online). For example, within the cold beverages category, items recorded included: regular soft drink, diet soft drink, energy drink, sports drink, flavoured milk, fruit juice, iced tea, water, and alcohol.

Where an item that was advertised also included branding for a particular food store, this was also noted. Advertisements for food products and food stores were not mutually exclusive (e.g. an advertisement by convenience store chain ‘7-Eleven’ may also promote food items such as doughnuts or hot coffee).

Additional information recorded included the location of the advertisements (e.g. train station, bus shelter or tram shelter). The total

number of bus and tram stops with shelters was counted (including those with no food advertisements) during the audit process so that the percentage of stops containing food advertisements could be calculated.

Data collectionData was collected between May and June 2013. To ensure every public transit location was included, all public transport stops (train stations, tram stops and bus stops) within the selected suburbs were identified using the Melways street directory.26 In addition, when auditing the suburbs, bus and tram routes were followed to capture any stops not marked in the street directory. Data was collected at all train stations and only at tram or bus stops when a shelter was present, as stops where no opportunity for advertising was available were not relevant to the study.

The auditing was undertaken by a single person (PS) who visited each of the identified transit stops. Prior to data collection, all authors were involved in the piloting of the audit tool and a rigorous process to verify its accuracy and reliability. Many of the advertisements were straight forward to record (e.g. diet soft drink at a bus shelter) however if ambiguity remained, photos or further details of the advertisement were recorded and a decision on how to code was undertaken by all authors.

The population density for each suburb was obtained from the 2011 Census of Population and Housing.27 The mean number of advertisements per square kilometre and per 1,000 residents was calculated for each suburb and examined by area-level disadvantage.

Data analysisStata v12.128 was used for all data analysis. The estimated marginal mean of the number of advertisements per suburb by area-level disadvantage was calculated with adjustments made for suburb population, geographic area and distance to the Melbourne central business district (Table 1). The proportion of stops with an advertisement present was calculated for each mode of transport, i.e., for train stations, tram and bus shelters (Table 2). The proportion of food advertisements in each product category was calculated (Figure 1). As cold beverages was the most prominent category, the proportion of specific items advertised within this category was also assessed (Figure 2). All advertisements branded by a food store included a product item and thus the food store category does not appear on Figure 1, as this would mean the categories would not be mutually exclusive and proportions would not total 100%. Instead, breakdown of the food store types that appear on the advertisements are presented in Figure 3. For each of the proportion estimates in Table 2 and in Figures 1–3, the equality of proportions tests were used to estimate the statistical difference (p<0.05) by area-level disadvantage.

Results

A total of 233 food advertisements were identified at the 558 public transit stops audited across the 20 sampled suburbs. Least-disadvantaged suburbs had a higher mean number of advertisements per suburb (13.6) compared to the most-disadvantaged

Table 1: Mean number of food advertisements according to area-level disadvantage, suburb population size and area.

Overall Least disadvantaged (n. advertisements = 136)

Most disadvantaged (n. advertisements = 97)

Suburb mean (s.d.) Suburb mean (95% CI) Suburb mean (95% CI) p-value

Number of advertisements (n 233) 11.7 (12.9) 13.6 (2.6–24.6) 9.7 (4.2–15.1) 0.514

Suburb population 17,269 (6,877) 16,145 (12,025–20,265) 18,393 (13,340–23,447) 0.480

Number of advertisements per suburb adjusted for suburb population

14.5 (6.8–22.1) 8.8 (1.2–16.6) 0.330

Suburb area (sq km) 6.8 (3.9) 5.5 (3.3–7.7) 8.0 (5.3–10.8) 0.148

Number of advertisements per suburb adjusted for suburb areas

12.7 (4.3–21.1) 10.6 (2.2–19.0) 0.741

Distance to CBD (km) 11.8 (7.0) 10.9 (7.6–14.2) 12.7 (6.9–18.4) 0.579

Number of advertisements per suburb adjusted for distance to CBD

13.0 (5.1–20.9) 10.3 (2.4–18.1) 0.635

Socioeconomic Status and Health Outdoor food advertising at public transit stops

416 Australian and New Zealand Journal of Public Health 2014 vol. 38 no. 5© 2014 Public Health Association of Australia

suburbs (9.7), although this difference was not statistically significant (Table 1). Similarly, no significant difference in the number of advertisements according to area-level disadvantage was observed after accounting for suburb population, geographic area and distance to the Melbourne CBD.

Across all types of public transit stops, the percentage with an advertisement was similar by area-level disadvantage; however, differences were observed by the type of stop (Table 2). A higher proportion of train stations in the least-disadvantaged suburbs had at least one advertisement present (86% vs. 42%, p=0.003). Conversely, fewer tram shelters in the least-disadvantaged areas featured food advertisements (32% vs. 50%. p=0.055). The proportion of bus stop shelters with food advertisements was similar in the least- and most-disadvantaged suburbs (22% and 25% respectively, p=0.547).

The product categories advertised did not differ significantly by area-level disadvantage with the exception of hot beverage items (13% in least-disadvantaged suburbs and 2% in most-disadvantaged, p=0.004), see Figure 1. Items categorised as cold beverages were more often advertised than items in any other food category. In the least-disadvantaged suburbs, 35% of all food advertisements were for cold beverage items while 45% of advertisements in most-disadvantaged suburbs were for cold beverages (p=0.121). Similar numbers of advertisements for snack foods were found in the least-disadvantaged (17%) and most-disadvantaged (16%) suburbs. No advertisements for cereal or fruit/vegetable items were observed.

Figure 2 shows that within the cold beverages category, advertisements for diet soft drink were more frequently observed in the least-disadvantaged areas compared to the most-disadvantaged areas (57% vs. 25%, p=0.002). Interestingly, no advertisements were observed for sugar-sweetened varieties of soft drinks. Advertisements for flavoured milk (8% vs. 25%, p=0.028) and fruit juice (2% vs. 20%, p=0.004) were less often seen in the least-disadvantaged areas. Advertisements for alcohol did not differ between areas (24% least-disadvantaged vs. 20% most-disadvantaged, p=0.642).

A number of food advertisements included branding for particular food stores (Figure 3). Convenience store advertisements were mostly for 7-Eleven stores, with the percentage of advertisements for 7-Eleven and the percentage of advertisements for all convenience stores both being higher in the least-disadvantaged suburbs compared to the most-disadvantaged suburbs (7-Eleven: 31% vs. 12%, p=0.045; total convenience stores: 37% vs. 15%, p=0.026). Conversely, advertisements for each of the three fast food chains represented were most often observed in the most-disadvantaged suburbs. In total, 48% of all advertisements featuring a particular store in the most-disadvantaged areas featured fast food chains, with the corresponding figure being 19% in the least-disadvantaged areas (p=0.003). Advertisements featuring

Figure 1: Percentage of each product category advertised by area-level disadvantage (n 233).

Figure 2: Percentage of specific cold beverage item advertised by area-level disadvantage (n 92).

Figure 3: Percentage of food store advertisements by area-level disadvantage (n 95).

Settle, Cameron and Thornton Article

2014 vol. 38 no. 5 Australian and New Zealand Journal of Public Health 417© 2014 Public Health Association of Australia

Socioeconomic Status and Health Outdoor food advertising at public transit stops

food stores other than fast food chains or convenience stores were generally promoting local restaurants and the percentage of these did not differ significantly by area-level disadvantage (p=0.498).

Discussion

This audit of outdoor food advertising at public transit stops within a sample of Melbourne suburbs found that the total number of advertisements did not differ between the least- and most-disadvantaged suburbs even after accounting for suburb population, area and distance to the CBD. We did, however, reveal socioeconomic patterning in the types of advertisements, with advertisements for fast food chains, flavoured milk and fruit juice all more common in more disadvantaged areas, and advertisements for hot beverages (tea, coffee), diet soft drinks and convenience stores more frequent in the least-disadvantaged areas.

Previous research has reported that outdoor food advertising usually promotes unhealthy foods and/or that unhealthy food advertising is more clustered in deprived communities.10,11,17,18,20-22 The findings from this study generally support the notion that outdoor food advertising in Melbourne does not promote more nutritious foods. Further, it was found that chain-brand fast food and diet soft drink options varied according to area-level disadvantage, which is of particular interest. It is well established that disadvantaged neighbourhoods have a higher number of fast food outlets compared to the least-disadvantaged neighbourhoods29 and findings from our study show the promotion of these stores is also more prominent in disadvantaged areas. Similar, in line with previous work that demonstrates the high amount of snack foods and soft drinks on display within Melbourne supermarkets,30,31 advertising of such items is also prominent. A higher percentage of unhealthy food advertisements in the most-disadvantaged suburbs contributes to potential inequalities in the obesogenic food environment6 with advertising considered to be part of the information environment.32 Such a food environment entrenches population norms around unhealthy eating. While the higher amount of unhealthy food advertising in the more disadvantaged suburbs works against the promotion of healthy eating in neighbourhoods with the

greatest need, the absence of advertising for healthy foods suggests there is room for improvement in all areas.

Public transit stops are important advertising locations as they are visited by large numbers of people each day and may therefore have a significant impact on food choices.22,33 This study found that 30% of public transit stops had at least one advertisement for a food item. Outdoor food advertising is of policy relevance at both state and federal level. VicHealth has previously recognised the importance of train stations as inclusive areas that contribute to community wellbeing33 while the House of Representatives Standing Committee on Social Policy and Legal Affairs has tabled a report on the regulation of these advertisements.24 The latter report raises concerns that outdoor food and beverage advertising may counteract public health campaigns.

The findings of this study may support and encourage government agencies to develop health promotion initiatives aimed at outdoor food advertising, either by promoting healthier food or potentially restricting advertising of unhealthy foods. The precise impact of changing the food advertising environment on food purchasing and consumption is yet to be determined. The results of this descriptive research should be seen as a necessary but not sufficient step toward the development of policy recommendations.

Strengths and limitations This is the first study we are aware of to examine outdoor food advertisements at all forms of public transit stops and to sample by area-level disadvantage. The analysis considered potential confounding factors that might influence the presence of outdoor food advertisements including suburb population

and area, and distance to the CBD. This study was limited to suburbs within a 20 km radius of the Deakin University Burwood campus and was not a random selection from across the Melbourne metropolitan area. Despite this limitation, there is no reason to assume any difference found between the least- and most-disadvantaged neighbourhoods in this study region would be different from other areas of Melbourne. Larger number of advertisements were observed at the train stations in the least-disadvantaged neighbourhoods, and this is likely to have been due to the presence of larger train stations (i.e. with more platforms) in these areas. The number of platforms at train stations was not counted in this study and should be considered in future research. The types of foods advertised is likely to be seasonal and data collection took place over one month in winter. Therefore, this data may not represent the type of outdoor food advertised in Melbourne at other times of the year. A final point to note is that by restricting our measures to only transit stops we have no indication of the total amount of outdoor food advertising within suburbs and whether observed associations would remain if all other outdoor advertisements (e.g. advertisements attached to shopping centres) were included.

Conclusion

This exploration of the prevalence of outdoor food advertising at Melbourne transit stops found a high proportion of stops contained advertisements for foods and beverages not considered nutritious with some variation in products advertised by area-level disadvantage. The findings are an important first step to quantify and raise awareness of unhealthy environmental exposures.

Table 2: Percentage of public transport stops with food advertisements by area-level disadvantage

Overall Least disadvantaged Most disadvantaged P values

Number of public transit stops audited 558 250 308

% of public transit stops with at least one advertisement 30.1% 30.4% (24.7–36.1) 29.9% (24.8–35.0) 0.892

Train stations audited (n) 41 22 19

% of train stations with at least one advertisement 65.9% 86.4% (71.2–100.0) 42.1% (18.6–65.6) 0.003

Bus shelters audited (n) 402 163 239

% of bus shelters with an advertisement 23.6% 22.1% (15.7–28.5) 24.7% (19.2–30.2) 0.547

Tram shelters audited (n) 115 65 50

% of tram shelters with an advertisement 40.0% 32.3% (20.1–43.7) 50% (36.1–63.9) 0.055

418 Australian and New Zealand Journal of Public Health 2014 vol. 38 no. 5© 2014 Public Health Association of Australia

Acknowledgements

The collection of data within Melbourne was supported by funding through Deakin University. LET is supported by a Deakin University Alfred Deakin Postdoctoral Research Fellowship. AJC is supported by a National Health and Medical Research Council Fellowship (#1013313).

References1. World Health Organization. Diet, Nutrition, and the

Prevention of Chronic Diseases. WHO Technical Report Series No.: 916. Geneva (CHE): WHO; 2003.

2. Australian Bureau of Statistics. 4364.0.55.001 - Australian Health Survey: First Results, 2011-12. Canberra (AUST): ABS; 2012.

3. King T, Kavanagh AM, Jolley D, Turrell G, Crawford D. Weight and place: a multilevel cross-sectional survey of area-level social disadvantage and overweight/obesity in Australia. Int J Obes. 2006;30(2):281-7.

4. Ni Mhurchu C, Vandevijvere S, Waterlander W, Thornton L, Kelly B, Cameron A, et al. Monitoring the availability of healthy and unhealthy foods and non‐alcoholic beverages in community and consumer retail food environments globally. Obes Rev. 2013;14(S1):108-19.

5. Papas MA, Alberg AJ, Ewing R, Helzlsouer KJ, Gary TL, Klassen AC. The built environment and obesity. Epidemiol Rev. 2007;29(1):129-43.

6. Story M, Kaphingst KM, Robinson-O’Brien R, Glanz K. Creating healthy food and eating environments: Policy and environmental approaches. Annu Rev Public Health. 2008;29:253-72.

7. Chandon P, Wansink B. Does food marketing need to make us fat? A review and solutions. Nutr Rev. 2012;70(10):571-93.

8. Harris JL, Bargh JA, Brownell KD. Priming effects of television food advertising on eating behavior. Health Psychol. 2009;28(4):404-13.

9. Cairns G, Angus K, Hastings G, Caraher M. Systematic reviews of the evidence on the nature, extent and effects of food marketing to children. A retrospective summary. Appetite. 2013;62:209-15.

10. Kelly B, Cretikos M, Rogers K, King L. The commercial food landscape: Outdoor food advertising around primary schools in Australia. Aust N Z J Public Health. 2008;32(6):522-8.

11. Maher A, Wilson N, Signal L. Advertising and availability of ‘obesogenic’foods around New Zealand secondary schools: A pilot study. N Z Med J. 2005;118(1218):U1556.

12. Zimmerman FJ. Using marketing muscle to sell fat: The rise of obesity in the modern economy. Ann Rev Public Health. 2011;32:285-306.

13. Kearney J. Food consumption trends and drivers. Philos Trans R Soc Lond B Biol Sci. 2010;365(1554):2793-807.

14. Lobstein T, Dibb S. Evidence of a possible link between obesogenic food advertising and child overweight. Obes Rev. 2005;6(3):203-8.

15. Henderson VR, Kelly B. Food advertising in the age of obesity: Content analysis of food advertising on general market and African American television. J Nutr Educ Behav. 2005;37(4):191-6.

16. Walton M, Pearce J, Day P. Examining the interaction between food outlets and outdoor food advertisements with primary school food environments. Health Place. 2009;15(3):841-8.

17. Adams J, Ganiti E, White M. Socio-economic differences in outdoor food advertising in a city in Northern England. Public Health Nutr. 2011;14(6):945.

18. Hillier A, Cole BL, Smith TE, Yancey AK, Williams JD, Grier SA, et al. Clustering of unhealthy outdoor advertisements around child-serving institutions: A comparison of three cities. Health Place. 2009;15(4): 935-45.

19. Kwate NO, Lee TH. Ghettoizing outdoor advertising: Disadvantage and ad panel density in black neighborhoods. J Urban Health. 2007;84(1):21-31.

20. Lesser LI, Zimmerman FJ, Cohen DA. Outdoor advertising, obesity, and soda consumption: A cross-sectional study. BMC Public Health. 2013;13(1):20.

21. Yancey AK, Cole BL, Brown R, Williams JD, Hillier A, Kline RS, et al. A cross‐sectional prevalence study of ethnically targeted and general audience outdoor obesity‐related advertising. Milbank Q. 2009;87(1):155-84.

22. Kelly B, Flood VM, Bicego C, Yeatman H. Derailing healthy choices: An audit of vending machines at train stations in NSW. Health Promot J Austr. 2012;23(1):73-5.

23. Pierce HL, Stafford JM, Daube M. The extent of alcohol advertising in Australia: An audit of bus stop advertisements. Med J Aust. 2013;198(9):478-9.

24. House of Representatives Standing Committee on Social Policy and Legal Affairs. Reclaimng Public Space. Inquiry into the Regulation of Billboard and Outdoor Advertsing [Internet]. Canberra (AUST): Parliament of the Commmonwealth of Australia; 2011 [cited 2013 Jul 21]. Available from: http://www.aph.gov.au/parliamentary_business/committees/house_of_representatives_committees?url=spla/outdoor%20advertising/report.htm

25. Australian Bureau of Statistics. What is SEIFA [Internet]. Canberra (AUST): ABS; 2013 [cited 2013 Apr 25]. Available from: http://www.abs.gov.au/

26. Melway. Greater Melbourne Street Directory. Melbourne (AUST): Melway Publishing; 2012.

27. Australian Bureau of Statistics. Census of Population and Housing 2011[Internet]. Canberra (AUST): ABS; 2013 [cited 2013 Aug 24]. Available from: http://www.abs.gov.au/census

28. STATA: statistical software [computer program]. Version 12.1. College Station (TX): Stata Corporation; 2013.

29. Fraser LK, Edwards KL, Cade J, Clarke GP. The geography of fast food outlets: A review. Int J Environ Res Public Health. 2010;7:2290-308.

30. Cameron AJ, Thornton LE, McNaughton SA, Crawford D. Variation in supermarket exposure to energy-dense snack foods by socioeconomic position. Public Health Nutr. 2012;1(1):1-8.

31. Thornton LE, Cameron AJ, McNaughton SA, Worsley A, Crawford DA. The availability of snack food displays that may trigger impulse purchases in Melbourne supermarkets. BMC Public Health. 2012;12:194.

32. Glanz K, Sallis JF, Saelens BE, Frank LD. Healthy nutrition environments: Concepts and measures. Am J Health Promot. 2005;19(5):330-3.

33. Village Well. Train Stations as Places for Community Wellbeing [Internet]. Melbourne (AUST): VicHealth; 2006 [cited 2013 Jul 21]. Available from: http://www.vichealth.vic.gov.au/Publications/Planning-Healthy-Environments/Train-Stations-as-Places-for-Community-Wellbeing.aspx

Supporting Information

Additional supporting information may be found in the online version of this article:

Supplementary File 1: Food product categories and specific food items recorded.

Settle, Cameron and Thornton Article