Food marketing in recreational sport settings in Canada: a cross-sectional audit in different policy environments using the Food and beverage Marketing Assessment Tool for Settings (FoodMATS)

Food marketing in recreational sport settings in Canada: a cross-sectional audit in different... Background: Children’s recreational sport settings typically sell energy dense, low nutrient products; however, it is unknown whether the same types of food and beverages are also marketed in these settings. Understanding food marketing in sports settings is important because the food industry often uses the promotion of physical activity to justify their products. This study aimed to document the ‘exposure’ and ‘power’ of food marketing present in public recreation facilities in Canada and assess differences between provinces with and without voluntary provincial nutrition guidelines for recreation facilities. Methods: Food marketing was measured in 51 sites using the Food and beverage Marketing Assessment Tool for Settings (FoodMATS). The frequency and repetition (‘exposure’) of food marketing and the presence of select marketing techniques, including child-targeted, sports-related, size, and healthfulness (‘power’), were assessed. Differences in ‘exposure’ and ‘power’ characteristics between sites in three guideline provinces (n =34) and a non-guideline province (n = 17) were assessed using Pearson’s Chi squared tests of homogeneity and Mann-Whitney U tests. Results: Ninety-eight percent of sites had food marketing present. The frequency of food marketing per site did not differ between guideline and non-guideline provinces (median =29; p = 0.576). Sites from guideline provinces had a significantly lower proportion of food marketing occasions that were “Least Healthy” (47.9%) than sites from the non-guideline province (73.5%; p < 0.001). Use of child-targeted and sports-related food marketing techniques was significantly higher in sites from guideline provinces (9.5% and 10.9%, respectively), than in the non-guideline province (1.9% and 4.5% respectively; p values < 0.001). It was more common in the non-guideline province to use child-targeted and sports-related techniques to promote “Least Healthy” items (100.0% and 68.4%, respectively), compared to the guideline provinces (59.3% and 52.0%, respectively). (Continued on next page) * Correspondence: kim.raine@ualberta.ca School of Public Health, University of Alberta, Edmonton, Alberta, Canada Centre for Health and Nutrition, University of Alberta, Edmonton, Alberta, Canada Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 2 of 11 (Continued from previous page) Conclusions: Recreation facilities are a source of children’s exposure to unhealthy food marketing. Having voluntary provincial nutrition guidelines that recommend provision of healthier foods was not related to the frequency of food marketing in recreation facilities but was associated with less frequent marketing of unhealthy foods. Policy makers should provide explicit food marketing regulations that complement provincial nutrition guidelines to fulfill their ethical responsibility to protect children and the settings where children spend time. Keywords: Food marketing, Recreational sport settings, Children and youth, Policy, Healthy eating Background the most important and feasible interventions to pro- Increased prevalence of childhood obesity is believed to mote children’s health [9]. In this regard, several Canad- be the product of “small, cumulative environmental ian provinces [Alberta (AB), British Columbia (BC), changes that have altered children’s physical activity and Nova Scotia (NS)] have introduced voluntary nutrition dietary patterns” (p.e1) [1]. By providing opportunities to guidelines to encourage healthier food provision in re- be active, recreation and sport facilities may be ideal creation facilities [10–12]. Guidelines introduced in 2015 sites to support childhood obesity prevention. Recreation in NS, Canada discouraged unhealthy food promotion, and sport facilities, defined as public or private commu- sponsorship, and marketing [12]. Taking a different nity centres that offer opportunities for physical activity approach, guidelines in AB, Canada, revised in 2012, rec- and programming for children and adults at a fee, have a ommended marketing healthier foods through competi- mandate to promote health and wellbeing [2]. However, tive pricing and placement [11]. Guidelines in the this mandate may be undermined by the unhealthy foods Canadian province of BC, revised in 2014, did not men- they offer [3] which are commonly deep fried foods, hot tion food marketing [10]. Even without specific food dogs, and sugary snacks and drinks [4]. In a systematic marketing recommendations, food marketing environ- review by Nelson et al. [5], no difference in children’s ments may improve in parallel with improved food weights was found between those who participated in provision as guidelines are implemented in recreation fa- extracurricular physical activity and those who did not, cilities. Once a new food product introduced into a re- in spite of the former being more physically active than creation facility, marketing may be used to increase the latter. Increased availability, marketing and con- consumers’“recognition, appeal and/or consumption” sumption of fast foods and soft drinks in sport settings [13] (p.9) of the product through pricing, placement, or may have contributed to this weight discrepancy [5]. promotion [14]. Thus, we aimed to investigate the differ- Food and beverage marketing (henceforth food mar- ence in food marketing environments between provinces keting) in recreation and sport facilities may influence with and without provincial nutrition guidelines. food attitudes, preferences and behaviors. A scoping re- Describing the nature and extent of food marketing in view of the relationship between watching sports and sport settings is a current gap in the literature [6]. The population health concluded that sport spectating may limited available research focuses on the prevalence of increase unhealthy eating behaviours from exposure to sport sponsorship [15] or testing the impact of experimen- unhealthy food sponsorship [6]. Unhealthy food market- tal food marketing techniques in recreation facilities on ing that uses sport or physical activity appeals is con- food choices [16, 17]. It is necessary to understand the cerning due to its associated impacts on product breadth, intensity, and characteristics of food marketing in likeability and nutritional quality. In a cross-sectional recreation facilities to inform healthy food policy and re- study of 10–14 year olds who participated in sports at a duce children’s exposure to unhealthy food marketing. local club in Australia, over two-thirds could recall at Marketing policies that reduce ‘exposure’ to and ‘power’ least one food and beverage company sponsor of their of food and beverage marketing are recommended by the club and 59% “liked to return the favour to these spon- WHO [18] and could reduce the impact of unhealthy food sors by buying their products” (p.4) [7]. Furthermore, marketing on children’s eating behaviors. both adults and children may experience a ‘halo effect’ To fill the gap in the literature regarding food market- when food is marketed with physical activity themes, ing in recreation facilities, this study aimed to document leading to more positive reactions and perceptions of the food and beverage marketing in public recreation product healthfulness [8]. and sport facilities in Canada and assess differences in Restricting unhealthy food and beverage sport spon- food marketing environments between facilities from sorship and improving healthy food availability in recre- provinces with voluntary nutrition guidelines and facil- ation and sport facilities have been ranked as some of ities from a province with no guidelines. This type of Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 3 of 11 investigation is valuable as it may reveal how well Data collection current nutrition guidelines designed to enhance healthy A trained EPL provincial coordinator or research as- food provision also protect (or do not protect) children sistant conducted observational audits using the Food from unhealthy food marketing. We aimed to explore Marketing Assessment Tool for Settings (FoodMATS) the ‘exposure’ (frequency, repetition) to and ‘power’ [19] between November 2015 and May 2016. The (healthfulness, child-targeting, sports-related, size) of FoodMATS captures the presence of food marketing food marketing in public recreation facilities. We hy- in recreation facilities, what food products, brands, pothesized that recreation facilities in provinces with and retailers were marketed, and whether persuasive voluntary nutrition guidelines would have less unhealthy (powerful) marketing techniques were used. At each food marketing (related to a difference in food provision) site, a trained rater photographed and recorded the but did not have any other a priori hypotheses for other following on the FoodMATS: outcomes assessed due to limited research that currently exists on this topic.  the frequency of food and beverage marketing in sports areas, food areas (concessions), and other Methods areas (entrance, hallways, parking lot), Setting and participants  the product, brand, or food retailer marketed, This study was part of a larger Eat Play Live (EPL) re-  whether the marketing occasion targeted children, search project evaluating the impact of voluntary provin-  whether the marketing occasion was related to cial nutrition guidelines on recreation and sport facility sports, and food environments including food availability, market-  the physical size of the marketing occasion. ing, and policy in Canada. Public recreation facilities in three provinces with existing provincial nutrition guide- One marketing occasion was defined as one adver- lines for recreation facilities (BC [10], AB [11], and NS tisement, promotion, or message (e.g. one sign), that [12]) and one province without provincial nutrition is intended to increase the “recognition, appeal and/ guidelines [Ontario (ON)] were included in the current or consumption” of a food or beverage products, study. Eligible facilities were those that provided food brands,orretailer[13] (p.9). Marketing occasions that services through vending or concession (such as a can- were not physical signage (e.g. product placement and teen, snack bar, café, or restaurant), had not made major pricing promotions) were counted but were not changes to their food environment since 2010, were will- assessed for targeting children, being related to ing and able to make changes to their food environment, sports, or their size as that would usually require and had year-round sport programming. reviewing product packaging which was beyond the Facilities were recruited for EPL between August 2015 scope of this study. and April 2016 by provincial parks and recreation organi- After each site visit, one registered dietitian (RD) zations and the EPL team. A buffer of 150 km (adjusted (RJLP) classified all marketing occasions according to by provinces if appropriate in regards to geography and their healthfulness using composite rankings (“Most budget) was used to identify a subsample of facilities near Healthy”, “Less Healthy”, “Least Healthy”) (Table 1) in- universities (n = 286) that were followed-up by telephone. formed by provincial nutrition guidelines [10–12]. Clas- Only 216 facilities were eligible to participate and 49 facil- sifications were checked by a second RD (KDR). We ities (22.7%) agreed to participate. Of the remaining, 141 calculated the repetition of food marketing in each site, did not respond to the invitation; 11 refused without rea- defined as the number of products, brands, or retailers son; 15 refused due to insufficient staff capacity (n =11), that were marketed at least three times per site. A Food- uninterested in research (n = 2), risk of being a control site MATS score was derived for each site based on the ‘ex- (n = 1), worried about competition (n = 1). Non-response posure’ to food and beverage marketing (defined as the greatly varied by province (ON 25%; BC 36%; AB 63%; NS frequency and repetition), and the ‘power’ of each mar- 92%). Two facilities had two separate buildings which we keting occasion (defined as the persuasiveness of mar- treated as individual sites for a total of 51 sites where food keting represented by its unhealthfulness, use of child- and beverage marketing was measured. Thirty-four sites targeted and/or sports-related techniques, and size). Our were from the three guideline provinces; 17 sites were definitions of exposure and power were operationalized from the one non-guideline province. A sample size of 43 from the WHO’s Exposure and Power of Marketing Mes- was required for the EPL project to detect a medium to sages model where exposure was explained as “the reach large effect (d = 0.8) in unhealthy food and beverage avail- and frequency of the marketing message”, and power ability in vending machines between two groups with α was “the creative content, design and execution of the =0.05. See methods for post hoc power analyses of the marketing message” [13] (p.11). Scores could range from sample size to detect change in marketing scores. zero to infinity with higher scores representing sites with Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 4 of 11 Table 1 Classification of Marketing Occasions by Healthfulness [19] Type “Most Healthy”“Less Healthy”“Least Healthy” a b Products / Brands Unprocessed foods and beverages Foods and beverages with some Processed energy-dense, nutrient- with no added fat, sugar or salt added fat, sugar, or salt poor items with high levels of fat, sugar, or salt Retailers Grocery stores, farmers’ markets Sit-down restaurants, cafeterias, Pizza, burger, taco, fried chicken, Sandwich outlets, smoothie coffee outlets, prepared grocery Asian, and ice cream outlets, pubs, outlets, salad bars stores, supplement stores lounges, alcohol stores Other All nutrition education or healthy None None eating promotion defined as a tangible food or beverage [14]), defined as a name or symbol that represents the maker of a product [14]), defined as a place where food can be purchased (store, restaurant, etc.) greater exposure to food marketing, along with more Data analysis powerful food marketing. FoodMATS data were entered, cleaned, and scored in The FoodMATS was previously validated by assessing Microsoft Excel 2013. Statistical analysis was completed correlations with recreation facility sponsorship and ad- wiht Statistical Package for the Social Sciences Version vertising dollars, and whether FoodMATS scores predict 23 (SPSS Inc., Chicago, IL, USA) was used with p<0.05 unhealthy food and beverage sales [19]. During pilot test- indicating statistical significance. Medians and interquar- ing the FoodMATS demonstrated very good to excellent tile ranges were used to describe the frequency and repe- inter-rater reliability (κ =0.88–1.00, p < 0.001; ICC = 0.97, tition of marketing, and FoodMATS scores. The p <0.001) [19]. prevalence of powerful features (healthfulness, child- Detailed methods on EPL and the FoodMATS have targeted, sports-related, size) was described using pro- been previously reported [19]. portions. Crosstabs were used to assess whether market- We also assessed post hoc whether food marketing ing occasions that used child-targeted and sports-related was related to the types of foods available for cus- marketing techniques differed by healthfulness. tomers to purchase (as opposed to any alternative Differences between guideline and non-guideline prov- such as the food marketing was related to sponsor- inces were assessed using Pearson’s Chi squared tests of ship or fundingprovidedtothe site by an outside homogeneity. Ordinal variables were collapsed into di- organization) by identifying “in house” products, chotomous groups to improve stability. Healthfulness was brands, and retailers. Products and brands were con- grouped into “Most Healthy”/“Less Healthy” versus “Least sidered “in house” if they were sold in vending ma- Healthy” as the latter are recommended to be restricted or chines or concessions within the site the marketing not available in recreation facilities [10–12]. Size was was found. Food retailers were considered “in house” grouped into small/medium versus large. Effect sizes are if they sold food or beverages within the site. Audits reported as Phi coefficients interpreted as 0.1 for small ef- conducted at concessions and in vending machines fects, 0.3 for medium effects, and 0.5 for large effects [20]. and product sales reports collected for the EPL study Due to unequal variances and non-normality, Mann- were used to check whether a product or brand was Whitney U tests were used to test differences between sold onsite. Names of concessions recorded in the guideline and non-guideline provinces for food marketing FoodMATS were used to determine if the marketed frequency, repetition, and FoodMATS scores. Post hoc food retailer was onsite. The classification was com- power analyses with G*Power (v3.1) revealed that our pleted by a trained graduate research assistant and sample size would have 73% chance of detecting a large ef- checked by RJLP. This type of classification may be fect (D = 0.80, t =2.01, α = 0.05) when using Mann- important to understand how food marketing is in- Whitney tests to compare mean ranks between two fluenced across different operational areas in the fa- groups, and assuming two-tailed normal distribution with cility, which may require different interventions if an α = 0.05; but would be insufficient to detect medium (D = association is found. For example, if most marketing 0.50, α =0.36) or small (D = 0.2, α =0.099) effect sizes. is for foods and beverages available onsite then food service operators may be the target of interventions. Results On the other hand, if there is marketing from out- Characteristics of guideline and non-guideline sites side retailers or for products/brands not sold within The majority of guideline (n = 23, 67.6%) and non- the facility, then an intervention may need to target guideline (n = 15, 88.2%) sites had one concession. Eight management or financial departments that contract sites in the guideline provinces had no concession(s) (23. out advertising space. 5%). Zero sites in the non-guideline province had no Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 5 of 11 concession(s). All other sites in guideline provinces (n =5, remaining food marketing occasions promoted food re- 14.7%) and the non-guideline province (n =2, 11.8%) had tailers (22.5%) or were nutrition education or general two or more concessions. Thirty-one guideline sites (91. healthy eating promotions (2.2%), such as government, 2%) and all 17 non-guideline sites (100.0%) had snack industry, or site developed posters that provided nutri- and/or beverage vending machines. Almost two-thirds of tion information or highlighted healthy food choices. sites in the guideline provinces (n = 22, 64.7%) and non- Most products (97.1%) and brands (85.8%) marketed guideline province (n = 11, 64.7%) had between one and were “in house”, but only 12.7% of marketing occasions four sports areas (see Table 2 for types of sports areas). for food retailers were “in house”. Food retailers that did One site in the guideline provinces had spaces for com- not sell food within the facility were promoted almost munity events such as dances but no dedicated sport eight times more often than “in house” food retailers. area. All other sites in the guideline (n = 11, 32.4%) and non-guideline provinces (n =6, 35.3%) had five or more Repetition sports areas. Overall, sites marketed a median of two products, Food marketing was present in all but one site (n = 50, brands, or retailers three or more times. However, the 98.0%), located in a guideline province. Most sites had food top quartile of sites repeatedly marketed between marketing in their food (concession) area(s) (n =41 out of three and 13 products, brands, and retailers at least 43 sites with concessions, 95.3%), sports area(s) (n =35 out three times within their site. There was no difference of 50 sites with sports areas, 70.0%), and other area(s) (n = in the number of repeated products, brands, and re- 46 out of 51 sites, 90.2%). Presence of food marketing dif- tailers between guideline and non-guideline provinces fered between sport area types, ranging from 2.6% of gym- (p = 0.217) (Table 5). nasiums to 81.3% of arenas having food marketing (Table 2). No single use courts, cycling studios, climbing areas, or Power other areas contained food marketing (Table 2). There were statistically significant differences in the proportions of food marketing occasions that were Exposure “Least Healthy”, child-targeted, sports-related, and Frequency large size between sites in guideline and non- A total of 1740 food marketing occasions were recorded guideline provinces (Table 5). across all sites. The frequency of promotions by location can be found in Table 3. Overall, the median number of Healthfulness of marketing food marketing occasions per site was 29 (IQR 13, 42) Overall, more than half of all food marketing occasions (Table 4). There was no statistical difference between the were considered “Least Healthy” (55.6%) (Table 5). There number of food marketing occasions between provinces was a significantly greater proportion of “Least Healthy” with and without guidelines (p = 0.576) (Table 4). food marketing occasions in the non-guideline province Products or brands were most frequently marketed, compared to the guideline province (X (1, N = 1740) comprising 75.3% of all marketing occasions. The =63.604, Phi coefficient = − 0.191, p < 0.001) (Table 5). Table 2 Number and proportion of sports areas with food marketing present (n = 188) All sites Guideline sites Non-guideline sites Sports area n Proportion of sports areas with n Proportion of sports areas with n Proportion of sports areas with food marketing present (%) food marketing present (%) food marketing present (%) All sports areas 188 36.2 119 34.5 69 41.2 Arenas 64 81.3 30 83.3 34 79.4 Fields 7 71.4 5 80.0 2 50.0 Tracks 4 25.0 3 66.7 1 0.0 Weight/Cardio room 24 25.0 19 31.6 5 0.0 Pool 24 12.5 16 18.8 8 0.0 Gymnasiums 38 2.6 32 3.1 6 0.0 Single-use courts 12 0.0 4 0.0 8 0.0 Cycle studios 6 0.0 5 0.0 1 0.0 Rock climbing walls 1 0.0 0 0.0 1 0.0 Other sport areas 8 0.0 5 0.0 3 0.0 Includes: indoor playground (n = 2), gymnastics area (n = 2), shuffle board (n = 1), ballet studio (n = 1), bowling alley (n = 1), skateboarding area (n =1) Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 6 of 11 Table 3 Number and proportion of food marketing occasions found in food, sports, and other area by type (n = 1740) Food (concession) n Proportion of all Sports areas n Proportion of all Other areas n Proportion of all areas food marketing in food marketing in food marketing in e e e food areas (%) sports areas (%) other areas (%) Checkout 229 30.8 Playing area 200 39.3 Indoor walls/ floors 70 14.4 Price promotions 159 21.3 Seating area 96 18.9 Facility TVs 24 4.9 c d Signs/ displays/ 150 20.2 Other 59 11.6 Other 22 4.5 table tents Menus 102 13.7 Scoreboard/clocks 44 8.6 Outdoor walls, 14 2.9 windows, doors Other 101 13.6 Change/locker rooms 15 2.9 Welcome desk 14 2.9 Outdoor signs, 10 2.1 furniture Facility pamphlets 10 2.1 Bathrooms 3 0.6 Vending machines 3 0.4 Vending machines in 61 12.0 Vending machines 320 65.7 spectator area Vending machines in 34 6.7 athlete area Total 744 100.0 Total 509 100.0 Total 487 100 Includes multiple pricing promotion types: combos; small versus regular portions; and healthy entrees, salads, beverages, and snacks versus regular; and other pricing. No supersize, all-you-can-eat, free refills, loyalty programs were found Includes marketing/branding on fridges, coolers, machines, garbage cans, recycling cans, menus, clocks etc. Includes marketing/branding on stairs, coolers, floors, bulletin boards, etc. Includes marketing on sandwich boards/posters Percentages may not add up to 100.0 due to rounding Child-targeted food marketing Sports-related food marketing Approximately, one in every 14 food marketing occa- Approximately 1 in every 11 food marketing occasions sions (7.2%) was targeted at children (Table 5). There (8.9%) were sports-related (Table 5). There was a was a significantly greater proportion of child-targeted significantly greater proportion of sports-related food food marketing occasions in guideline provinces than in marketing occasions in guideline provinces than in the 2 2 non-guideline provinces (X (1, N = 1377) =25.817, Phi non-guideline province (X (1, N = 1377) =14.923, p <0. coefficient = 0.137, p < 0.001) (Table 5). 001, Phi coefficient = 0.086) (Table 5). Across all sites, the healthfulness of food marketing Overall, 52.0% of all sports-related food marketing oc- occasions targeted at children and not targeted at chil- casions were “Least Healthy” (n = 64); however, it was dren were similar, however, 100.0% of the food market- more common in non-guideline sites with 68.4% (n = 51) ing occasions targeted at children in non-guideline to have sports-related food marketing occasions for provinces were “Least Healthy” (n = 8), compared to only “Least Healthy” products, brands, or retailers compared 59.3% in guideline provinces (n = 54) (Fig. 1). to 49.0% (n = 53) in guideline sites (Fig. 2). Table 4 Exposure to food and beverage marketing occasions for facility areas for guideline and non-guideline provinces (n = 1740) All sites (n = 51) Guideline sites (n = 34) Non-guideline sites (n = 17) a a a b Median IQR Median IQR Median IQR P value Frequency of food marketing occasions (n) Total Site 29.0 13.0, 42.0 28.5 5.5, 42.3 29.0 20.0, 42.5 p = 0.576 Food Areas 13.0 7.3, 20.8 15.0 5.0, 25.0 12.0 7.5, 17.0 p = 0.447 Sports Areas 5.5 0.0, 13.0 6.0 0.0, 15.0 5.0 2.0, 12.5 p = 0.787 Other Areas 7.0 3.0, 13.0 7.0 3.0, 13.0 11.0 3.5, 15.5 p = 0.389 Repetition of food marketing occasions (n) Total Site 2.0 1.0, 3.0 2.0 1.0, 3.0 2.0 1.0, 3.0 p = 0.217 Interquartile Range (IQR) = 25th percentile, 75th percentile asymptotic significance (2-tailed) from Mann-Whitney test difference of mean ranks between scores Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 7 of 11 Table 5 Power of food and beverage marketing occasions for guideline and non-guideline provinces (n = 1740) Power feature All sites (n = 51) Guideline sites (n = 34) Non-guideline sites (n = 17) n (missing) % n (missing) % n (missing) % P value Healthfulness n = 1740 (0) n = 1212 (0) n = 528 (0) p < 0.001 Most Healthy 420 24.1 358 29.5 62 11.7 Less Healthy 352 20.2 274 22.6 78 14.8 Least Healthy 968 55.6 580 47.9 388 73.5 Child-targeted n = 1377 (5) n = 953 (4) n = 424 (1) p < 0.001 Targeted at children 99 7.2 91 9.5 8 1.9 Sports-related n = 1377 (5) n = 953 (4) n = 424 (1) p < 0.001 Related to sports 123 8.9 104 10.9 19 4.5 Size total n = 1375 (6) n = 952 (4) n = 423 (2) p = 0.001 Small 444 32.3 282 29.6 162 38.3 Medium 257 18.7 193 20.3 64 15.1 Large 674 49.0 477 50.1 197 46.6 a 2 asymptotic significance (2-sided) from Chi tests for homogeneity evidence of animated or fictional characters, taste appeals, humour, action-adventure, fantasy, fun shapes or colours, competitions, give-aways, cartoonish font, or used a child actor to advertise a food or beverage product/brand that would appeal to children (Prowse et al. submitted to IJBNPA November 2017, IJBN-D-17-00585) any reference to physical activity, exercise, sport, game, recreation, performance or competition, a design feature relevant to sport settings (Prowse et al. submitted to IJBNPA November 2017, IJBN-D-17-00585) small: less than one 8.5 × 11 in. paper (Prowse et al. submitted to IJBNPA November 2017, IJBN-D-17-00585) outdoor medium: one to ten 8.5 × 11 in. paper(s); indoor medium: one to three 8.5 × 11 in. paper(s) (Prowse et al. submitted to IJBNPA November 2017, IJBN-D-17-00585) outdoor large: more than ten 8.5 × 11 in. paper(s); indoor large: more than three- 8.5 × 11 in. paper(s) (Prowse et al. submitted to IJBNPA November 2017, IJBN-D-17-00585) Size of marketing FoodMATS scores Almost half of all food marketing occasions were large Overall, the median score was 43.3 (IQR 18.6, 71.0) with and one-third were small (Table 5). There was a signifi- higher scores indicating greater exposure to food market- cantly greater proportion of large food marketing occa- ing, along with more powerfulfood marketing. There was sions in the guideline province than the non-guideline no statistically significant difference in FoodMATS scores provinces (X (2, N = 1375) =11.718, Phi coefficient = 0. between guideline (median = 42.7, IQR 4.6, 70.1) and non- 092, p = 0.003) (Table 5). guideline provinces (median = 43.3, IQR 29.5, 71.5). Fig. 1 Distribution by healthfulness for child-targeted and non-child-targeted marketing occasions in guideline and non-guideline provinces (N = 1377) Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 8 of 11 Fig. 2 Distribution by healthfulness for sports-related and non-sports-related marketing occasions in guideline and non-guideline provinces (N =1377) Discussion The lack of difference in FoodMATS scores may high- Food marketing, such as signs, posters, branding, pricing light that there are multiple components to food market- promotions, and product placement, was found to be ing that need to be considered in policy interventions. present in almost all recreation facility sites with un- Current provincial nutrition guidelines incompletely ad- healthy products, brands, or retailers marketed on more dress food marketing by merely recommending what than half of the occasions. Our study found mixed re- product should or should not marketed (i.e. healthy ver- sults in differences between sites in guideline and non- sus unhealthy food) which is only one component of guideline provinces, differing by ‘power’ (healthfulness of marketing strategies. How foods and beverages are mar- food marketing, targeting children, using appeals of keted (targeted to children, sports-related, and physical physical activity, and having large signs) but not by ‘ex- size, as well as potentially other characteristics not posure’ (frequency, and repetition) nor FoodMATS assessed in this study) should also be regulated in order scores (the composite of ‘power’ and ‘exposure’). to protect children from exposure to powerful food mar- It may be surprising that the FoodMATS scores did not keting. That being said, protecting children’s environ- differ between guideline types despite differences in ments from all unhealthy food marketing would reduce ‘power’. This null result may be related to the fact that we children’s exposure to food marketing and thus make could only use a non-parametric test to compare mean discussions regarding other powerful features redundant. ranks. If actual values were assessed, findings may have No previous research has evaluated food marketing in shown a difference since the 25th percentile of FoodMATS sports settings as comprehensively as this study. Carter score is almost 25 points (84.4%) lower in the guideline et al. [21] identified 131 food and beverage companies provinces than in the non-guideline province. Secondly, that advertised on sports clubs’ websites in New Zea- FoodMATS scores were calculated by assessing each com- land. Although we did not measure the number of dif- ponent of ‘power’ individually rather than cumulatively. If ferent marketers, we found that only a couple products, ‘power’ was scored based on the cumulative presence of brands, and/or retailers were marketed repeatedly in a marketing techniques, the FoodMATS scores in the non- site. The findings from both Carter et al. [21] and this guideline province may have been higher since more mar- study suggest that there are several food industry actors keting occasions that used child-targeted and sports- involved in food marketing in recreation and sport related techniques were for “Least Healthy” products, mak- facilities. Kelly et al. [22] found that sports club food ing it easier to see differences between guideline types. sponsors in Australia most commonly provided jersey However, our approach of evaluating each component in- branding (53% of sponsors), official partnership (52%), dividually proposes the idea that the impact of food mar- recognition in club newsletters (29%), signs (28%), and keting on children’s food preferences and behaviours may onsite availability of sponsors’ product (24%). This pro- remain unchanged if one marketing technique is replaced ject also found that signage was a common marketing by another (e.g. replace sports-related food marketing oc- channel and that most products marketed were available casions with child-targeted food marketing occasions). for purchase in the facility. However, the marketing Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 9 of 11 techniques and channels captured by Kelly et al. [22] marketing. Although it did not measure sponsorship only overlap to a limited extent with the FoodMATS specifically, it captured a breadth of marketing ap- since Kelly et al. [22] only evaluated sponsorship and the proaches the food industry uses in sport settings. FoodMATS broadly assessed food marketing within multiple areas of the facility including concessions and vending machines. The breadth of food marketing found Implications & recommendations in this study suggests that sponsorship may be only one To our knowledge, this is the first study to investigate of many strategies the food industry uses to market their the ‘exposure’ and ‘power’ of food marketing in sport product, brand, or retailer in sports settings. settings, a place where children gather that should be The proportion of marketing occasions that were “Least free from unhealthy food marketing [18], and to examine Healthy” (55%) found in this study is similar to the pro- differences in food marketing environments according to portion of food sponsors classified as unhealthy by Carter presence of regional voluntary nutrition guidelines. We et al. [21] (using the New Zealand Food and Beverage found differences between what and how foods and bev- Classification System) and Kelly et al. [22] (through expert erages were marketed, but not in the frequency or repe- consensus classification). These consistent findings sug- tition of marketing. Findings suggest that the presence gest that food marketing environments in recreation and of voluntary provincial nutrition guidelines that focus on sport facilities are not health promoting. what food provision rather than food marketing may be The greater use of child-targeted marketing in the insufficient to impact the frequency of marketing but guideline province may reflect that the provincial guide- may influence the healthfulness of marketing. It is pos- lines tend to focus on improving children’s environments sible that provincial nutrition guidelines improve the and may be related to efforts by sites from guideline foods available for sale onsite which impacts their mar- provinces to move towards offering and promoting keting. However, nutrition guidelines for food provision healthier options for children. It could also be explained can only be expected to go so far; a study of food pro- by other factors that we did not assess including differ- motions in public schools in Vancouver, Canada found ences in the prevalence of onsite child programming or that almost one-quarter of promotions were for “Choose proximity of schools to the recreation facility. Least” and “Not Recommended” foods and beverages The difference in sports-related marketing between [24] even though provincial school nutrition guidelines guideline and non-guideline provinces is surprising be- there discouraged unhealthy food marketing (e.g. post- cause the prevalence of sports areas with food marketing ers, coupons, and branded equipment) [25]. was lower in the guideline provinces than in the non- The presence of unhealthy food marketing found in guideline province, and the number of sports areas was schools by Velazquez et al. [24] and in recreation facil- similarly distributed in both groups. The study did find ities presented here despite the presence of nutrition that food marketing was variable depending on the type guidelines suggests that it should not be assumed that of sport, consistent with previous research [21, 23]. Des- healthy food provision policies will translate to healthier pite this, it is unclear whether differences in sport types food promotion. On the other hand, it may also be between sites in guideline and non-guideline provinces shortsighted to assume that food provision policies will explains the different prevalence of sports-related food have no impact on food marketing within its applicable marketing between guideline types. setting. Although child-targeted marketing techniques were Strengths and limitations used infrequently, recreation and sport facilities still The results of this study must be interpreted cautiously offer multiple exposures to unhealthy food marketing. due to its cross-sectional design and small, non- Regardless of their power, children will likely still see representative sample; yet, this is the largest known as- such marketing and be impacted by it. Sport sponsorship sessment of food marketing in recreation facilities in is not inherently child-targeted, but a study of 5–12 year Canada. Unfortunately, our small sample size did not olds in New Zealand found that 76% of children can allow us to investigate whether differences in marketing correctly match sponsors to their respective sport [26]. environments existed in sites between guideline prov- Pettigrew et al. [26] also found that even when children inces in relation to their variable food marketing recom- mismatched sponsors with sports, 83% of children se- mendations. Similarly, we had insufficient power to lected an unhealthy food brand for that sport, sug- adjust for clustering effects within provinces resulting in gesting that children have a strong association of confidence intervals narrower than if we could have ad- unhealthy food with sport. A photo-based project in justed for clustering. Despite its limitations, the Food- New Zealand revealed that 83% of beverages 10– MATS is a theoretically grounded reliable validated tool 12 year olds associate with sport were not consistent that provides broad and detailed information on food with dietary guidelines [27]. Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 10 of 11 The presence of unhealthy food marketing in almost all features were occasionally present. Having provincial nu- recreation facilities studied in Canada is worrisome from a trition guidelines did not appear to impact the frequency population health perspective. Thousands of children, or repetition of food marketing in recreation facilities, youth, and families use public recreation facilities in but was associated with less unhealthy food promotion, Canada [28, 29], thus the reach of food marketing is including the products marketed with child-targeted or broad. Kelly et al. [23] estimated that Australian children sports-related techniques. As researchers and practi- may be exposed up to 64,000 person-hours of food and tioners work to improve food environments in sport set- beverage sponsorships per week depending on the sport. tings, targeting food marketing as an environmental It is not reasonable to expect recreation facilities that sell factor appears important for supporting healthy eating. food to be free of food marketing (although food sponsor- ship may be unnecessary), but marketing environments Acknowledgements The authors would like to thank the EPL provincial coordinators, Dona Tomlin, could be improved to be less pervasive across recreation Susan Caswell, and Sherry Jarvis, for their assistance with data collection. and sport facilities and be used to promote healthy prod- ucts only. Marketing policies that reduce ‘exposure’ to and Funding ‘power’ of food and beverage marketing are recommended EPL received specific funding from the Heart & Stroke Foundation of Canada. by the WHO [18] and could reduce the impact of un- RP is supported by a Canadian Institute for Health Research (CIHR) Frederick Banting and Charles Best Canada Graduate Scholarship - Doctoral Award, healthy food marketing on children’s eating behaviors. In- and a Women and Children Health Research Initiative Graduate Studentship stitutions, such as recreation facilities, may consider funded by the Stollery Children’s Hospital Foundation. VC is supported by a generating food marketing restrictions to complement CIHR New Investigator Salary Award. SFLK was supported in part from a CIHR funded Canada Research Chair. LCM received salary support from the BC food provision policies s in order to more comprehen- Children’s Hospital Research Institute. sively promote healthy diets [30]. Future research should explore the relationships of Availability of data and materials food marketing in children’s sport settings with other The datasets used and/or analysed during the current study are available from environmental factors (food availability, food sales) and the corresponding author on reasonable request. the impact of food marketing in sport settings on indi- vidual and population diet and health outcomes. Investi- Authors’ contributions PJN, LCM, SFLK, and KDR contributed to the design of the project. RJLP gating the impact of food marketing according to developed the FoodMATS tool. RJLP collected the data with help from other FoodMATS scores may help to understand how to re- EPL provincial coordinators (see acknowledgements). RLJP analyzed and duce the impact of food marketing by identifying ideal interpreted data. RJLP developed the manuscript with critical editorial support from DLO, VC, LCM, KA, SFLK, PJN, and KDR. All authors read and food marketing scores and generating strong, specific approved the final manuscript. recommendations for policymakers to restrict unhealthy food marketing and sponsorship in children’s sport set- Authors’ information tings. Researchers should consider assessing differences R.P. is a registered dietitian and a PhD Candidate. in food marketing between sport types (hockey versus soccer), facility type (public versus private funding; sin- Ethics approval and consent to participate gle versus multi-sport), competition levels, and commu- Ethics approval was obtained for the EPL project at all participating universities (Universities of British Columbia, Victoria, Alberta, and Waterloo, and Dalhousie nities in which these facilities are located (high versus University). low income; urban versus rural).Such research may re- veal whether certain populations are at greater risk of Competing interests exposure to unhealthy food marketing environments. The authors declare that they have no competing interests. Understanding such differences could identify where to focus interventions to have the greatest population im- Publisher’sNote pact on diet, health, and childhood obesity. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Conclusions Author details It is argued that the food industry often overemphasizes School of Public Health, University of Alberta, Edmonton, Alberta, Canada. the importance of physical activity deliberately [31, 32] 2 School of Exercise Science, Physical and Health Education, University of to “[deflect] attention from its possible role in the obes- Victoria, Victoria, British Columbia, Canada. Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, ity epidemic” (p.244) [33]. The overwhelming presence Alberta, Canada. Faculty of Physical Education and Recreation, University of of food marketing in recreation facilities may be evi- 5 Alberta, Edmonton, Alberta, Canada. BC Children’s Hospital Research dence of one method used by the food industry to do so. Institute, Vancouver, British Columbia, Canada. School of Population and Public Health, University of British Columbia, Vancouver, British Columbia, Over half of food products, brands, and retailers mar- Canada. Healthy Populations Institute, Dalhousie University, Halifax, Nova keted in public recreation facilities were “Least Healthy”. 8 Scotia, Canada. Centre for Health and Nutrition, University of Alberta, Although not common, child-targeted and sports-related Edmonton, Alberta, Canada. Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 11 of 11 Received: 27 November 2017 Accepted: 10 April 2018 24. Velazquez CE, Black JL, Ahmadi N. Food and beverage promotions in Vancouver schools: a study of the prevalence and characteristics of in- school advertising, messaging, and signage. Prev Med Rep. 2015;2:757–64. 25. Government of British Columbia. Guidelines for food and beverage sales in B.C. schools. Victoria: Province of British Columbia; 2013. References 26. Pettigrew S, Rosenberg M, Ferguson R, Houghton S, Wood L. Game on: do 1. Brennan LK, Brownson RC, Orleans CT. Childhood obesity policy research children absorb sports sponsorship messages? Public Health Nutr. 2013;16: and practice: evidence for policy and environmental strategies. Am J Prev 2197–204. Med. 2014;46:e1–16. 27. Smith M, Jenkin G, Signal L, McLean R. Consuming calories and creating 2. Government of Alberta. Active Alberta 2011–2021. http://culture.alberta. cavities: beverages NZ children associate with sport. Appetite. 2014;81:209–17. ca/recreation/active-alberta/pdf/Active-Alberta-Policy.pdf.Accessed 24 28. Randall Conrad and Associates, Roma M. ARPA’s infrastructure committee: Nov 2017. operations survey summary report. Edmonton: Alberta Recreation and Parks 3. Olstad DL, Downs SM, Raine KD, Berry TR, McCargar LJ. Improving children’s Association; 2006. nutrition environments: a survey of adoption and implementation of 29. Naylor PJ, Wekken SV, Trill D, Kirbyson A. Facilitating healthier food nutrition guidelines in recreational facilities. BMC Public Health. 2011;11:423. environments in public recreation facilities: results of a pilot project in 4. Naylor PJ, Bridgewater L, Purcell M, Ostry A, Wekken SV. Publically funded British Columbia, Canada. J Park Recreat Adm. 2010;28:37–58. recreation facilities: obesogenic environments for children and families? Int 30. Prowse R. Food marketing to children in Canada: a settings-based scoping J Environ Res Public Health. 2010;7:2208–21. review on exposure, power and impact. Health Promot Chronic Dis Prev 5. Nelson TF, Stovitz SD, Thomas M, Lavoi NM, Bauer KW, Neumark-Sztainer D. Can. 2017;37:274–92. Do youth sports prevent pediatric obesity? A systematic review and 31. Brownell KD, Warner KE. The perils of ignoring history: big tobacco played commentary. Curr Sports Med Rep. 2011;10:360–70. dirty and millions died. How similar is big food? Milbank Q. 2009;87:259–94. 6. Inoue Y, Berg BK, Chelladurai P. Spectator sport and population health: a 32. Kirk SFL, Penney TL, Freedhoff Y. Running away with the facts on food and scoping study. J Sport Manage. 2015;29:705–25. fatness. Public Health Nutr. 2010;13:147–8. 7. Kelly B, Baur LA, Bauman AE, King L, Chapman K, Smith BJ. “Food company 33. Folta SC, Goldberg JP, Economos C, Bell R, Meltzer R. Food advertising sponsors are kind, generous and cool”: (mis)conceptions of junior sports targeted at school-age children: a content analysis. J Nutr Educ Behav. players. Int J Behav Nutr Phys Act. 2011;8:95. 2006;38:244–8. 8. Castonguay J. Sugar and sports age differences in children’s responses to a high sugar cereal advertisement portraying physical activities. Comm Res. 2015; https://doi.org/10.1177/0093650215587357. 9. Kelly B, King L, Bauman AE, Baur LA, Macniven R, Chapman K, et al. Identifying important and feasible policies and actions for health at community sports clubs: a consensus-generating approach. J Sci Med Sport. 2014;17:61–6. 10. Government of British Columbia. Healthier choices in vending machines in BC public buildings. Victoria: Province of British Columbia; 2014. 11. Alberta Health and Wellness. The Alberta nutrition guidelines for children and youth. Edmonton: Government of Alberta; 2010. 12. Government of Nova Scotia. Healthy eating in recreation and sport settings guidelines. 2015. 13. World Health Organization. A framework for implementing the set of recommendations on the marketing of foods and non-alcoholic beverages to children. 2012. 14. Lee NR, Kotler P. Social marketing: influencing behaviors for good. Thousand Islands: Sage; 2011. 15. Carter M-A, Edwards R, Signal L, Hoek J. Availability and marketing of food and beverages to children through sports settings: a systematic review. Public Health Nutr. 2012;15:1373–9. 16. Wolfenden L, Kingsland M, Rowland BC, Dodds P, Gillham K, Yoong SL, et al. Improving availability, promotion and purchase of fruit and vegetable and non sugar-sweetened drink products at community sporting clubs: a randomised trial. Int J Behav Nutr Phys Act. 2015;12:35. 17. Olstad DL, Goonewardene LA, McCargar LJ, Raine KD. Choosing healthier foods in recreational sports settings: a mixed methods investigation of the impact of nudging and an economic incentive. Int J Behav Nutr Phys Act. 2014;11:6. 18. World Health Organization: Set of recommendations on the marketing of foods and non-alcoholic beverages to children. 2010. 19. Prowse RJL, Naylor P-J, Olstad DL, Carson V, Masse L, Storey K, et al. Reliability and validity of novel tool to comprehensively assess food and beverage marketing in recreational sport settings. Int J Behav Nutr Phys Act. 2018. https://doi.org/10.1186/s12966-018-0667-3. 20. Cohen J. Statistical power analysis for the behavioral sciences. revised ed. New York: Academic Press; 1977. 21. Carter MA, Signal L, Edwards R, Hoek J, Maher A. Food, fizzy, and football: promoting unhealthy food and beverages through sport - a New Zealand case study. BMC Public Health. 2013;13:126. 22. Kelly B, Baur LA, Bauman AE, King L, Chapman K, Smith BJ. Food and drink sponsorship of children’s sport in Australia: who pays? Health Promt Int. 2010;26:188–95. 23. Kelly B, Bauman AE, Baur LA. Population estimates of Australian children’s exposure to food and beverage sponsorship of sports clubs. J Sci Med Sport. 2014;17:394–8. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Behavioral Nutrition and Physical Activity Springer Journals

Food marketing in recreational sport settings in Canada: a cross-sectional audit in different policy environments using the Food and beverage Marketing Assessment Tool for Settings (FoodMATS)

Free
11 pages
Loading next page...
 
/lp/springer_journal/food-marketing-in-recreational-sport-settings-in-canada-a-cross-IxJMKXvO0i
Publisher
BioMed Central
Copyright
Copyright © 2018 by The Author(s).
Subject
Medicine & Public Health; Clinical Nutrition; Behavioral Sciences; Health Promotion and Disease Prevention
eISSN
1479-5868
D.O.I.
10.1186/s12966-018-0673-5
Publisher site
See Article on Publisher Site

Abstract

Background: Children’s recreational sport settings typically sell energy dense, low nutrient products; however, it is unknown whether the same types of food and beverages are also marketed in these settings. Understanding food marketing in sports settings is important because the food industry often uses the promotion of physical activity to justify their products. This study aimed to document the ‘exposure’ and ‘power’ of food marketing present in public recreation facilities in Canada and assess differences between provinces with and without voluntary provincial nutrition guidelines for recreation facilities. Methods: Food marketing was measured in 51 sites using the Food and beverage Marketing Assessment Tool for Settings (FoodMATS). The frequency and repetition (‘exposure’) of food marketing and the presence of select marketing techniques, including child-targeted, sports-related, size, and healthfulness (‘power’), were assessed. Differences in ‘exposure’ and ‘power’ characteristics between sites in three guideline provinces (n =34) and a non-guideline province (n = 17) were assessed using Pearson’s Chi squared tests of homogeneity and Mann-Whitney U tests. Results: Ninety-eight percent of sites had food marketing present. The frequency of food marketing per site did not differ between guideline and non-guideline provinces (median =29; p = 0.576). Sites from guideline provinces had a significantly lower proportion of food marketing occasions that were “Least Healthy” (47.9%) than sites from the non-guideline province (73.5%; p < 0.001). Use of child-targeted and sports-related food marketing techniques was significantly higher in sites from guideline provinces (9.5% and 10.9%, respectively), than in the non-guideline province (1.9% and 4.5% respectively; p values < 0.001). It was more common in the non-guideline province to use child-targeted and sports-related techniques to promote “Least Healthy” items (100.0% and 68.4%, respectively), compared to the guideline provinces (59.3% and 52.0%, respectively). (Continued on next page) * Correspondence: kim.raine@ualberta.ca School of Public Health, University of Alberta, Edmonton, Alberta, Canada Centre for Health and Nutrition, University of Alberta, Edmonton, Alberta, Canada Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 2 of 11 (Continued from previous page) Conclusions: Recreation facilities are a source of children’s exposure to unhealthy food marketing. Having voluntary provincial nutrition guidelines that recommend provision of healthier foods was not related to the frequency of food marketing in recreation facilities but was associated with less frequent marketing of unhealthy foods. Policy makers should provide explicit food marketing regulations that complement provincial nutrition guidelines to fulfill their ethical responsibility to protect children and the settings where children spend time. Keywords: Food marketing, Recreational sport settings, Children and youth, Policy, Healthy eating Background the most important and feasible interventions to pro- Increased prevalence of childhood obesity is believed to mote children’s health [9]. In this regard, several Canad- be the product of “small, cumulative environmental ian provinces [Alberta (AB), British Columbia (BC), changes that have altered children’s physical activity and Nova Scotia (NS)] have introduced voluntary nutrition dietary patterns” (p.e1) [1]. By providing opportunities to guidelines to encourage healthier food provision in re- be active, recreation and sport facilities may be ideal creation facilities [10–12]. Guidelines introduced in 2015 sites to support childhood obesity prevention. Recreation in NS, Canada discouraged unhealthy food promotion, and sport facilities, defined as public or private commu- sponsorship, and marketing [12]. Taking a different nity centres that offer opportunities for physical activity approach, guidelines in AB, Canada, revised in 2012, rec- and programming for children and adults at a fee, have a ommended marketing healthier foods through competi- mandate to promote health and wellbeing [2]. However, tive pricing and placement [11]. Guidelines in the this mandate may be undermined by the unhealthy foods Canadian province of BC, revised in 2014, did not men- they offer [3] which are commonly deep fried foods, hot tion food marketing [10]. Even without specific food dogs, and sugary snacks and drinks [4]. In a systematic marketing recommendations, food marketing environ- review by Nelson et al. [5], no difference in children’s ments may improve in parallel with improved food weights was found between those who participated in provision as guidelines are implemented in recreation fa- extracurricular physical activity and those who did not, cilities. Once a new food product introduced into a re- in spite of the former being more physically active than creation facility, marketing may be used to increase the latter. Increased availability, marketing and con- consumers’“recognition, appeal and/or consumption” sumption of fast foods and soft drinks in sport settings [13] (p.9) of the product through pricing, placement, or may have contributed to this weight discrepancy [5]. promotion [14]. Thus, we aimed to investigate the differ- Food and beverage marketing (henceforth food mar- ence in food marketing environments between provinces keting) in recreation and sport facilities may influence with and without provincial nutrition guidelines. food attitudes, preferences and behaviors. A scoping re- Describing the nature and extent of food marketing in view of the relationship between watching sports and sport settings is a current gap in the literature [6]. The population health concluded that sport spectating may limited available research focuses on the prevalence of increase unhealthy eating behaviours from exposure to sport sponsorship [15] or testing the impact of experimen- unhealthy food sponsorship [6]. Unhealthy food market- tal food marketing techniques in recreation facilities on ing that uses sport or physical activity appeals is con- food choices [16, 17]. It is necessary to understand the cerning due to its associated impacts on product breadth, intensity, and characteristics of food marketing in likeability and nutritional quality. In a cross-sectional recreation facilities to inform healthy food policy and re- study of 10–14 year olds who participated in sports at a duce children’s exposure to unhealthy food marketing. local club in Australia, over two-thirds could recall at Marketing policies that reduce ‘exposure’ to and ‘power’ least one food and beverage company sponsor of their of food and beverage marketing are recommended by the club and 59% “liked to return the favour to these spon- WHO [18] and could reduce the impact of unhealthy food sors by buying their products” (p.4) [7]. Furthermore, marketing on children’s eating behaviors. both adults and children may experience a ‘halo effect’ To fill the gap in the literature regarding food market- when food is marketed with physical activity themes, ing in recreation facilities, this study aimed to document leading to more positive reactions and perceptions of the food and beverage marketing in public recreation product healthfulness [8]. and sport facilities in Canada and assess differences in Restricting unhealthy food and beverage sport spon- food marketing environments between facilities from sorship and improving healthy food availability in recre- provinces with voluntary nutrition guidelines and facil- ation and sport facilities have been ranked as some of ities from a province with no guidelines. This type of Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 3 of 11 investigation is valuable as it may reveal how well Data collection current nutrition guidelines designed to enhance healthy A trained EPL provincial coordinator or research as- food provision also protect (or do not protect) children sistant conducted observational audits using the Food from unhealthy food marketing. We aimed to explore Marketing Assessment Tool for Settings (FoodMATS) the ‘exposure’ (frequency, repetition) to and ‘power’ [19] between November 2015 and May 2016. The (healthfulness, child-targeting, sports-related, size) of FoodMATS captures the presence of food marketing food marketing in public recreation facilities. We hy- in recreation facilities, what food products, brands, pothesized that recreation facilities in provinces with and retailers were marketed, and whether persuasive voluntary nutrition guidelines would have less unhealthy (powerful) marketing techniques were used. At each food marketing (related to a difference in food provision) site, a trained rater photographed and recorded the but did not have any other a priori hypotheses for other following on the FoodMATS: outcomes assessed due to limited research that currently exists on this topic.  the frequency of food and beverage marketing in sports areas, food areas (concessions), and other Methods areas (entrance, hallways, parking lot), Setting and participants  the product, brand, or food retailer marketed, This study was part of a larger Eat Play Live (EPL) re-  whether the marketing occasion targeted children, search project evaluating the impact of voluntary provin-  whether the marketing occasion was related to cial nutrition guidelines on recreation and sport facility sports, and food environments including food availability, market-  the physical size of the marketing occasion. ing, and policy in Canada. Public recreation facilities in three provinces with existing provincial nutrition guide- One marketing occasion was defined as one adver- lines for recreation facilities (BC [10], AB [11], and NS tisement, promotion, or message (e.g. one sign), that [12]) and one province without provincial nutrition is intended to increase the “recognition, appeal and/ guidelines [Ontario (ON)] were included in the current or consumption” of a food or beverage products, study. Eligible facilities were those that provided food brands,orretailer[13] (p.9). Marketing occasions that services through vending or concession (such as a can- were not physical signage (e.g. product placement and teen, snack bar, café, or restaurant), had not made major pricing promotions) were counted but were not changes to their food environment since 2010, were will- assessed for targeting children, being related to ing and able to make changes to their food environment, sports, or their size as that would usually require and had year-round sport programming. reviewing product packaging which was beyond the Facilities were recruited for EPL between August 2015 scope of this study. and April 2016 by provincial parks and recreation organi- After each site visit, one registered dietitian (RD) zations and the EPL team. A buffer of 150 km (adjusted (RJLP) classified all marketing occasions according to by provinces if appropriate in regards to geography and their healthfulness using composite rankings (“Most budget) was used to identify a subsample of facilities near Healthy”, “Less Healthy”, “Least Healthy”) (Table 1) in- universities (n = 286) that were followed-up by telephone. formed by provincial nutrition guidelines [10–12]. Clas- Only 216 facilities were eligible to participate and 49 facil- sifications were checked by a second RD (KDR). We ities (22.7%) agreed to participate. Of the remaining, 141 calculated the repetition of food marketing in each site, did not respond to the invitation; 11 refused without rea- defined as the number of products, brands, or retailers son; 15 refused due to insufficient staff capacity (n =11), that were marketed at least three times per site. A Food- uninterested in research (n = 2), risk of being a control site MATS score was derived for each site based on the ‘ex- (n = 1), worried about competition (n = 1). Non-response posure’ to food and beverage marketing (defined as the greatly varied by province (ON 25%; BC 36%; AB 63%; NS frequency and repetition), and the ‘power’ of each mar- 92%). Two facilities had two separate buildings which we keting occasion (defined as the persuasiveness of mar- treated as individual sites for a total of 51 sites where food keting represented by its unhealthfulness, use of child- and beverage marketing was measured. Thirty-four sites targeted and/or sports-related techniques, and size). Our were from the three guideline provinces; 17 sites were definitions of exposure and power were operationalized from the one non-guideline province. A sample size of 43 from the WHO’s Exposure and Power of Marketing Mes- was required for the EPL project to detect a medium to sages model where exposure was explained as “the reach large effect (d = 0.8) in unhealthy food and beverage avail- and frequency of the marketing message”, and power ability in vending machines between two groups with α was “the creative content, design and execution of the =0.05. See methods for post hoc power analyses of the marketing message” [13] (p.11). Scores could range from sample size to detect change in marketing scores. zero to infinity with higher scores representing sites with Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 4 of 11 Table 1 Classification of Marketing Occasions by Healthfulness [19] Type “Most Healthy”“Less Healthy”“Least Healthy” a b Products / Brands Unprocessed foods and beverages Foods and beverages with some Processed energy-dense, nutrient- with no added fat, sugar or salt added fat, sugar, or salt poor items with high levels of fat, sugar, or salt Retailers Grocery stores, farmers’ markets Sit-down restaurants, cafeterias, Pizza, burger, taco, fried chicken, Sandwich outlets, smoothie coffee outlets, prepared grocery Asian, and ice cream outlets, pubs, outlets, salad bars stores, supplement stores lounges, alcohol stores Other All nutrition education or healthy None None eating promotion defined as a tangible food or beverage [14]), defined as a name or symbol that represents the maker of a product [14]), defined as a place where food can be purchased (store, restaurant, etc.) greater exposure to food marketing, along with more Data analysis powerful food marketing. FoodMATS data were entered, cleaned, and scored in The FoodMATS was previously validated by assessing Microsoft Excel 2013. Statistical analysis was completed correlations with recreation facility sponsorship and ad- wiht Statistical Package for the Social Sciences Version vertising dollars, and whether FoodMATS scores predict 23 (SPSS Inc., Chicago, IL, USA) was used with p<0.05 unhealthy food and beverage sales [19]. During pilot test- indicating statistical significance. Medians and interquar- ing the FoodMATS demonstrated very good to excellent tile ranges were used to describe the frequency and repe- inter-rater reliability (κ =0.88–1.00, p < 0.001; ICC = 0.97, tition of marketing, and FoodMATS scores. The p <0.001) [19]. prevalence of powerful features (healthfulness, child- Detailed methods on EPL and the FoodMATS have targeted, sports-related, size) was described using pro- been previously reported [19]. portions. Crosstabs were used to assess whether market- We also assessed post hoc whether food marketing ing occasions that used child-targeted and sports-related was related to the types of foods available for cus- marketing techniques differed by healthfulness. tomers to purchase (as opposed to any alternative Differences between guideline and non-guideline prov- such as the food marketing was related to sponsor- inces were assessed using Pearson’s Chi squared tests of ship or fundingprovidedtothe site by an outside homogeneity. Ordinal variables were collapsed into di- organization) by identifying “in house” products, chotomous groups to improve stability. Healthfulness was brands, and retailers. Products and brands were con- grouped into “Most Healthy”/“Less Healthy” versus “Least sidered “in house” if they were sold in vending ma- Healthy” as the latter are recommended to be restricted or chines or concessions within the site the marketing not available in recreation facilities [10–12]. Size was was found. Food retailers were considered “in house” grouped into small/medium versus large. Effect sizes are if they sold food or beverages within the site. Audits reported as Phi coefficients interpreted as 0.1 for small ef- conducted at concessions and in vending machines fects, 0.3 for medium effects, and 0.5 for large effects [20]. and product sales reports collected for the EPL study Due to unequal variances and non-normality, Mann- were used to check whether a product or brand was Whitney U tests were used to test differences between sold onsite. Names of concessions recorded in the guideline and non-guideline provinces for food marketing FoodMATS were used to determine if the marketed frequency, repetition, and FoodMATS scores. Post hoc food retailer was onsite. The classification was com- power analyses with G*Power (v3.1) revealed that our pleted by a trained graduate research assistant and sample size would have 73% chance of detecting a large ef- checked by RJLP. This type of classification may be fect (D = 0.80, t =2.01, α = 0.05) when using Mann- important to understand how food marketing is in- Whitney tests to compare mean ranks between two fluenced across different operational areas in the fa- groups, and assuming two-tailed normal distribution with cility, which may require different interventions if an α = 0.05; but would be insufficient to detect medium (D = association is found. For example, if most marketing 0.50, α =0.36) or small (D = 0.2, α =0.099) effect sizes. is for foods and beverages available onsite then food service operators may be the target of interventions. Results On the other hand, if there is marketing from out- Characteristics of guideline and non-guideline sites side retailers or for products/brands not sold within The majority of guideline (n = 23, 67.6%) and non- the facility, then an intervention may need to target guideline (n = 15, 88.2%) sites had one concession. Eight management or financial departments that contract sites in the guideline provinces had no concession(s) (23. out advertising space. 5%). Zero sites in the non-guideline province had no Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 5 of 11 concession(s). All other sites in guideline provinces (n =5, remaining food marketing occasions promoted food re- 14.7%) and the non-guideline province (n =2, 11.8%) had tailers (22.5%) or were nutrition education or general two or more concessions. Thirty-one guideline sites (91. healthy eating promotions (2.2%), such as government, 2%) and all 17 non-guideline sites (100.0%) had snack industry, or site developed posters that provided nutri- and/or beverage vending machines. Almost two-thirds of tion information or highlighted healthy food choices. sites in the guideline provinces (n = 22, 64.7%) and non- Most products (97.1%) and brands (85.8%) marketed guideline province (n = 11, 64.7%) had between one and were “in house”, but only 12.7% of marketing occasions four sports areas (see Table 2 for types of sports areas). for food retailers were “in house”. Food retailers that did One site in the guideline provinces had spaces for com- not sell food within the facility were promoted almost munity events such as dances but no dedicated sport eight times more often than “in house” food retailers. area. All other sites in the guideline (n = 11, 32.4%) and non-guideline provinces (n =6, 35.3%) had five or more Repetition sports areas. Overall, sites marketed a median of two products, Food marketing was present in all but one site (n = 50, brands, or retailers three or more times. However, the 98.0%), located in a guideline province. Most sites had food top quartile of sites repeatedly marketed between marketing in their food (concession) area(s) (n =41 out of three and 13 products, brands, and retailers at least 43 sites with concessions, 95.3%), sports area(s) (n =35 out three times within their site. There was no difference of 50 sites with sports areas, 70.0%), and other area(s) (n = in the number of repeated products, brands, and re- 46 out of 51 sites, 90.2%). Presence of food marketing dif- tailers between guideline and non-guideline provinces fered between sport area types, ranging from 2.6% of gym- (p = 0.217) (Table 5). nasiums to 81.3% of arenas having food marketing (Table 2). No single use courts, cycling studios, climbing areas, or Power other areas contained food marketing (Table 2). There were statistically significant differences in the proportions of food marketing occasions that were Exposure “Least Healthy”, child-targeted, sports-related, and Frequency large size between sites in guideline and non- A total of 1740 food marketing occasions were recorded guideline provinces (Table 5). across all sites. The frequency of promotions by location can be found in Table 3. Overall, the median number of Healthfulness of marketing food marketing occasions per site was 29 (IQR 13, 42) Overall, more than half of all food marketing occasions (Table 4). There was no statistical difference between the were considered “Least Healthy” (55.6%) (Table 5). There number of food marketing occasions between provinces was a significantly greater proportion of “Least Healthy” with and without guidelines (p = 0.576) (Table 4). food marketing occasions in the non-guideline province Products or brands were most frequently marketed, compared to the guideline province (X (1, N = 1740) comprising 75.3% of all marketing occasions. The =63.604, Phi coefficient = − 0.191, p < 0.001) (Table 5). Table 2 Number and proportion of sports areas with food marketing present (n = 188) All sites Guideline sites Non-guideline sites Sports area n Proportion of sports areas with n Proportion of sports areas with n Proportion of sports areas with food marketing present (%) food marketing present (%) food marketing present (%) All sports areas 188 36.2 119 34.5 69 41.2 Arenas 64 81.3 30 83.3 34 79.4 Fields 7 71.4 5 80.0 2 50.0 Tracks 4 25.0 3 66.7 1 0.0 Weight/Cardio room 24 25.0 19 31.6 5 0.0 Pool 24 12.5 16 18.8 8 0.0 Gymnasiums 38 2.6 32 3.1 6 0.0 Single-use courts 12 0.0 4 0.0 8 0.0 Cycle studios 6 0.0 5 0.0 1 0.0 Rock climbing walls 1 0.0 0 0.0 1 0.0 Other sport areas 8 0.0 5 0.0 3 0.0 Includes: indoor playground (n = 2), gymnastics area (n = 2), shuffle board (n = 1), ballet studio (n = 1), bowling alley (n = 1), skateboarding area (n =1) Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 6 of 11 Table 3 Number and proportion of food marketing occasions found in food, sports, and other area by type (n = 1740) Food (concession) n Proportion of all Sports areas n Proportion of all Other areas n Proportion of all areas food marketing in food marketing in food marketing in e e e food areas (%) sports areas (%) other areas (%) Checkout 229 30.8 Playing area 200 39.3 Indoor walls/ floors 70 14.4 Price promotions 159 21.3 Seating area 96 18.9 Facility TVs 24 4.9 c d Signs/ displays/ 150 20.2 Other 59 11.6 Other 22 4.5 table tents Menus 102 13.7 Scoreboard/clocks 44 8.6 Outdoor walls, 14 2.9 windows, doors Other 101 13.6 Change/locker rooms 15 2.9 Welcome desk 14 2.9 Outdoor signs, 10 2.1 furniture Facility pamphlets 10 2.1 Bathrooms 3 0.6 Vending machines 3 0.4 Vending machines in 61 12.0 Vending machines 320 65.7 spectator area Vending machines in 34 6.7 athlete area Total 744 100.0 Total 509 100.0 Total 487 100 Includes multiple pricing promotion types: combos; small versus regular portions; and healthy entrees, salads, beverages, and snacks versus regular; and other pricing. No supersize, all-you-can-eat, free refills, loyalty programs were found Includes marketing/branding on fridges, coolers, machines, garbage cans, recycling cans, menus, clocks etc. Includes marketing/branding on stairs, coolers, floors, bulletin boards, etc. Includes marketing on sandwich boards/posters Percentages may not add up to 100.0 due to rounding Child-targeted food marketing Sports-related food marketing Approximately, one in every 14 food marketing occa- Approximately 1 in every 11 food marketing occasions sions (7.2%) was targeted at children (Table 5). There (8.9%) were sports-related (Table 5). There was a was a significantly greater proportion of child-targeted significantly greater proportion of sports-related food food marketing occasions in guideline provinces than in marketing occasions in guideline provinces than in the 2 2 non-guideline provinces (X (1, N = 1377) =25.817, Phi non-guideline province (X (1, N = 1377) =14.923, p <0. coefficient = 0.137, p < 0.001) (Table 5). 001, Phi coefficient = 0.086) (Table 5). Across all sites, the healthfulness of food marketing Overall, 52.0% of all sports-related food marketing oc- occasions targeted at children and not targeted at chil- casions were “Least Healthy” (n = 64); however, it was dren were similar, however, 100.0% of the food market- more common in non-guideline sites with 68.4% (n = 51) ing occasions targeted at children in non-guideline to have sports-related food marketing occasions for provinces were “Least Healthy” (n = 8), compared to only “Least Healthy” products, brands, or retailers compared 59.3% in guideline provinces (n = 54) (Fig. 1). to 49.0% (n = 53) in guideline sites (Fig. 2). Table 4 Exposure to food and beverage marketing occasions for facility areas for guideline and non-guideline provinces (n = 1740) All sites (n = 51) Guideline sites (n = 34) Non-guideline sites (n = 17) a a a b Median IQR Median IQR Median IQR P value Frequency of food marketing occasions (n) Total Site 29.0 13.0, 42.0 28.5 5.5, 42.3 29.0 20.0, 42.5 p = 0.576 Food Areas 13.0 7.3, 20.8 15.0 5.0, 25.0 12.0 7.5, 17.0 p = 0.447 Sports Areas 5.5 0.0, 13.0 6.0 0.0, 15.0 5.0 2.0, 12.5 p = 0.787 Other Areas 7.0 3.0, 13.0 7.0 3.0, 13.0 11.0 3.5, 15.5 p = 0.389 Repetition of food marketing occasions (n) Total Site 2.0 1.0, 3.0 2.0 1.0, 3.0 2.0 1.0, 3.0 p = 0.217 Interquartile Range (IQR) = 25th percentile, 75th percentile asymptotic significance (2-tailed) from Mann-Whitney test difference of mean ranks between scores Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 7 of 11 Table 5 Power of food and beverage marketing occasions for guideline and non-guideline provinces (n = 1740) Power feature All sites (n = 51) Guideline sites (n = 34) Non-guideline sites (n = 17) n (missing) % n (missing) % n (missing) % P value Healthfulness n = 1740 (0) n = 1212 (0) n = 528 (0) p < 0.001 Most Healthy 420 24.1 358 29.5 62 11.7 Less Healthy 352 20.2 274 22.6 78 14.8 Least Healthy 968 55.6 580 47.9 388 73.5 Child-targeted n = 1377 (5) n = 953 (4) n = 424 (1) p < 0.001 Targeted at children 99 7.2 91 9.5 8 1.9 Sports-related n = 1377 (5) n = 953 (4) n = 424 (1) p < 0.001 Related to sports 123 8.9 104 10.9 19 4.5 Size total n = 1375 (6) n = 952 (4) n = 423 (2) p = 0.001 Small 444 32.3 282 29.6 162 38.3 Medium 257 18.7 193 20.3 64 15.1 Large 674 49.0 477 50.1 197 46.6 a 2 asymptotic significance (2-sided) from Chi tests for homogeneity evidence of animated or fictional characters, taste appeals, humour, action-adventure, fantasy, fun shapes or colours, competitions, give-aways, cartoonish font, or used a child actor to advertise a food or beverage product/brand that would appeal to children (Prowse et al. submitted to IJBNPA November 2017, IJBN-D-17-00585) any reference to physical activity, exercise, sport, game, recreation, performance or competition, a design feature relevant to sport settings (Prowse et al. submitted to IJBNPA November 2017, IJBN-D-17-00585) small: less than one 8.5 × 11 in. paper (Prowse et al. submitted to IJBNPA November 2017, IJBN-D-17-00585) outdoor medium: one to ten 8.5 × 11 in. paper(s); indoor medium: one to three 8.5 × 11 in. paper(s) (Prowse et al. submitted to IJBNPA November 2017, IJBN-D-17-00585) outdoor large: more than ten 8.5 × 11 in. paper(s); indoor large: more than three- 8.5 × 11 in. paper(s) (Prowse et al. submitted to IJBNPA November 2017, IJBN-D-17-00585) Size of marketing FoodMATS scores Almost half of all food marketing occasions were large Overall, the median score was 43.3 (IQR 18.6, 71.0) with and one-third were small (Table 5). There was a signifi- higher scores indicating greater exposure to food market- cantly greater proportion of large food marketing occa- ing, along with more powerfulfood marketing. There was sions in the guideline province than the non-guideline no statistically significant difference in FoodMATS scores provinces (X (2, N = 1375) =11.718, Phi coefficient = 0. between guideline (median = 42.7, IQR 4.6, 70.1) and non- 092, p = 0.003) (Table 5). guideline provinces (median = 43.3, IQR 29.5, 71.5). Fig. 1 Distribution by healthfulness for child-targeted and non-child-targeted marketing occasions in guideline and non-guideline provinces (N = 1377) Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 8 of 11 Fig. 2 Distribution by healthfulness for sports-related and non-sports-related marketing occasions in guideline and non-guideline provinces (N =1377) Discussion The lack of difference in FoodMATS scores may high- Food marketing, such as signs, posters, branding, pricing light that there are multiple components to food market- promotions, and product placement, was found to be ing that need to be considered in policy interventions. present in almost all recreation facility sites with un- Current provincial nutrition guidelines incompletely ad- healthy products, brands, or retailers marketed on more dress food marketing by merely recommending what than half of the occasions. Our study found mixed re- product should or should not marketed (i.e. healthy ver- sults in differences between sites in guideline and non- sus unhealthy food) which is only one component of guideline provinces, differing by ‘power’ (healthfulness of marketing strategies. How foods and beverages are mar- food marketing, targeting children, using appeals of keted (targeted to children, sports-related, and physical physical activity, and having large signs) but not by ‘ex- size, as well as potentially other characteristics not posure’ (frequency, and repetition) nor FoodMATS assessed in this study) should also be regulated in order scores (the composite of ‘power’ and ‘exposure’). to protect children from exposure to powerful food mar- It may be surprising that the FoodMATS scores did not keting. That being said, protecting children’s environ- differ between guideline types despite differences in ments from all unhealthy food marketing would reduce ‘power’. This null result may be related to the fact that we children’s exposure to food marketing and thus make could only use a non-parametric test to compare mean discussions regarding other powerful features redundant. ranks. If actual values were assessed, findings may have No previous research has evaluated food marketing in shown a difference since the 25th percentile of FoodMATS sports settings as comprehensively as this study. Carter score is almost 25 points (84.4%) lower in the guideline et al. [21] identified 131 food and beverage companies provinces than in the non-guideline province. Secondly, that advertised on sports clubs’ websites in New Zea- FoodMATS scores were calculated by assessing each com- land. Although we did not measure the number of dif- ponent of ‘power’ individually rather than cumulatively. If ferent marketers, we found that only a couple products, ‘power’ was scored based on the cumulative presence of brands, and/or retailers were marketed repeatedly in a marketing techniques, the FoodMATS scores in the non- site. The findings from both Carter et al. [21] and this guideline province may have been higher since more mar- study suggest that there are several food industry actors keting occasions that used child-targeted and sports- involved in food marketing in recreation and sport related techniques were for “Least Healthy” products, mak- facilities. Kelly et al. [22] found that sports club food ing it easier to see differences between guideline types. sponsors in Australia most commonly provided jersey However, our approach of evaluating each component in- branding (53% of sponsors), official partnership (52%), dividually proposes the idea that the impact of food mar- recognition in club newsletters (29%), signs (28%), and keting on children’s food preferences and behaviours may onsite availability of sponsors’ product (24%). This pro- remain unchanged if one marketing technique is replaced ject also found that signage was a common marketing by another (e.g. replace sports-related food marketing oc- channel and that most products marketed were available casions with child-targeted food marketing occasions). for purchase in the facility. However, the marketing Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 9 of 11 techniques and channels captured by Kelly et al. [22] marketing. Although it did not measure sponsorship only overlap to a limited extent with the FoodMATS specifically, it captured a breadth of marketing ap- since Kelly et al. [22] only evaluated sponsorship and the proaches the food industry uses in sport settings. FoodMATS broadly assessed food marketing within multiple areas of the facility including concessions and vending machines. The breadth of food marketing found Implications & recommendations in this study suggests that sponsorship may be only one To our knowledge, this is the first study to investigate of many strategies the food industry uses to market their the ‘exposure’ and ‘power’ of food marketing in sport product, brand, or retailer in sports settings. settings, a place where children gather that should be The proportion of marketing occasions that were “Least free from unhealthy food marketing [18], and to examine Healthy” (55%) found in this study is similar to the pro- differences in food marketing environments according to portion of food sponsors classified as unhealthy by Carter presence of regional voluntary nutrition guidelines. We et al. [21] (using the New Zealand Food and Beverage found differences between what and how foods and bev- Classification System) and Kelly et al. [22] (through expert erages were marketed, but not in the frequency or repe- consensus classification). These consistent findings sug- tition of marketing. Findings suggest that the presence gest that food marketing environments in recreation and of voluntary provincial nutrition guidelines that focus on sport facilities are not health promoting. what food provision rather than food marketing may be The greater use of child-targeted marketing in the insufficient to impact the frequency of marketing but guideline province may reflect that the provincial guide- may influence the healthfulness of marketing. It is pos- lines tend to focus on improving children’s environments sible that provincial nutrition guidelines improve the and may be related to efforts by sites from guideline foods available for sale onsite which impacts their mar- provinces to move towards offering and promoting keting. However, nutrition guidelines for food provision healthier options for children. It could also be explained can only be expected to go so far; a study of food pro- by other factors that we did not assess including differ- motions in public schools in Vancouver, Canada found ences in the prevalence of onsite child programming or that almost one-quarter of promotions were for “Choose proximity of schools to the recreation facility. Least” and “Not Recommended” foods and beverages The difference in sports-related marketing between [24] even though provincial school nutrition guidelines guideline and non-guideline provinces is surprising be- there discouraged unhealthy food marketing (e.g. post- cause the prevalence of sports areas with food marketing ers, coupons, and branded equipment) [25]. was lower in the guideline provinces than in the non- The presence of unhealthy food marketing found in guideline province, and the number of sports areas was schools by Velazquez et al. [24] and in recreation facil- similarly distributed in both groups. The study did find ities presented here despite the presence of nutrition that food marketing was variable depending on the type guidelines suggests that it should not be assumed that of sport, consistent with previous research [21, 23]. Des- healthy food provision policies will translate to healthier pite this, it is unclear whether differences in sport types food promotion. On the other hand, it may also be between sites in guideline and non-guideline provinces shortsighted to assume that food provision policies will explains the different prevalence of sports-related food have no impact on food marketing within its applicable marketing between guideline types. setting. Although child-targeted marketing techniques were Strengths and limitations used infrequently, recreation and sport facilities still The results of this study must be interpreted cautiously offer multiple exposures to unhealthy food marketing. due to its cross-sectional design and small, non- Regardless of their power, children will likely still see representative sample; yet, this is the largest known as- such marketing and be impacted by it. Sport sponsorship sessment of food marketing in recreation facilities in is not inherently child-targeted, but a study of 5–12 year Canada. Unfortunately, our small sample size did not olds in New Zealand found that 76% of children can allow us to investigate whether differences in marketing correctly match sponsors to their respective sport [26]. environments existed in sites between guideline prov- Pettigrew et al. [26] also found that even when children inces in relation to their variable food marketing recom- mismatched sponsors with sports, 83% of children se- mendations. Similarly, we had insufficient power to lected an unhealthy food brand for that sport, sug- adjust for clustering effects within provinces resulting in gesting that children have a strong association of confidence intervals narrower than if we could have ad- unhealthy food with sport. A photo-based project in justed for clustering. Despite its limitations, the Food- New Zealand revealed that 83% of beverages 10– MATS is a theoretically grounded reliable validated tool 12 year olds associate with sport were not consistent that provides broad and detailed information on food with dietary guidelines [27]. Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 10 of 11 The presence of unhealthy food marketing in almost all features were occasionally present. Having provincial nu- recreation facilities studied in Canada is worrisome from a trition guidelines did not appear to impact the frequency population health perspective. Thousands of children, or repetition of food marketing in recreation facilities, youth, and families use public recreation facilities in but was associated with less unhealthy food promotion, Canada [28, 29], thus the reach of food marketing is including the products marketed with child-targeted or broad. Kelly et al. [23] estimated that Australian children sports-related techniques. As researchers and practi- may be exposed up to 64,000 person-hours of food and tioners work to improve food environments in sport set- beverage sponsorships per week depending on the sport. tings, targeting food marketing as an environmental It is not reasonable to expect recreation facilities that sell factor appears important for supporting healthy eating. food to be free of food marketing (although food sponsor- ship may be unnecessary), but marketing environments Acknowledgements The authors would like to thank the EPL provincial coordinators, Dona Tomlin, could be improved to be less pervasive across recreation Susan Caswell, and Sherry Jarvis, for their assistance with data collection. and sport facilities and be used to promote healthy prod- ucts only. Marketing policies that reduce ‘exposure’ to and Funding ‘power’ of food and beverage marketing are recommended EPL received specific funding from the Heart & Stroke Foundation of Canada. by the WHO [18] and could reduce the impact of un- RP is supported by a Canadian Institute for Health Research (CIHR) Frederick Banting and Charles Best Canada Graduate Scholarship - Doctoral Award, healthy food marketing on children’s eating behaviors. In- and a Women and Children Health Research Initiative Graduate Studentship stitutions, such as recreation facilities, may consider funded by the Stollery Children’s Hospital Foundation. VC is supported by a generating food marketing restrictions to complement CIHR New Investigator Salary Award. SFLK was supported in part from a CIHR funded Canada Research Chair. LCM received salary support from the BC food provision policies s in order to more comprehen- Children’s Hospital Research Institute. sively promote healthy diets [30]. Future research should explore the relationships of Availability of data and materials food marketing in children’s sport settings with other The datasets used and/or analysed during the current study are available from environmental factors (food availability, food sales) and the corresponding author on reasonable request. the impact of food marketing in sport settings on indi- vidual and population diet and health outcomes. Investi- Authors’ contributions PJN, LCM, SFLK, and KDR contributed to the design of the project. RJLP gating the impact of food marketing according to developed the FoodMATS tool. RJLP collected the data with help from other FoodMATS scores may help to understand how to re- EPL provincial coordinators (see acknowledgements). RLJP analyzed and duce the impact of food marketing by identifying ideal interpreted data. RJLP developed the manuscript with critical editorial support from DLO, VC, LCM, KA, SFLK, PJN, and KDR. All authors read and food marketing scores and generating strong, specific approved the final manuscript. recommendations for policymakers to restrict unhealthy food marketing and sponsorship in children’s sport set- Authors’ information tings. Researchers should consider assessing differences R.P. is a registered dietitian and a PhD Candidate. in food marketing between sport types (hockey versus soccer), facility type (public versus private funding; sin- Ethics approval and consent to participate gle versus multi-sport), competition levels, and commu- Ethics approval was obtained for the EPL project at all participating universities (Universities of British Columbia, Victoria, Alberta, and Waterloo, and Dalhousie nities in which these facilities are located (high versus University). low income; urban versus rural).Such research may re- veal whether certain populations are at greater risk of Competing interests exposure to unhealthy food marketing environments. The authors declare that they have no competing interests. Understanding such differences could identify where to focus interventions to have the greatest population im- Publisher’sNote pact on diet, health, and childhood obesity. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Conclusions Author details It is argued that the food industry often overemphasizes School of Public Health, University of Alberta, Edmonton, Alberta, Canada. the importance of physical activity deliberately [31, 32] 2 School of Exercise Science, Physical and Health Education, University of to “[deflect] attention from its possible role in the obes- Victoria, Victoria, British Columbia, Canada. Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, ity epidemic” (p.244) [33]. The overwhelming presence Alberta, Canada. Faculty of Physical Education and Recreation, University of of food marketing in recreation facilities may be evi- 5 Alberta, Edmonton, Alberta, Canada. BC Children’s Hospital Research dence of one method used by the food industry to do so. Institute, Vancouver, British Columbia, Canada. School of Population and Public Health, University of British Columbia, Vancouver, British Columbia, Over half of food products, brands, and retailers mar- Canada. Healthy Populations Institute, Dalhousie University, Halifax, Nova keted in public recreation facilities were “Least Healthy”. 8 Scotia, Canada. Centre for Health and Nutrition, University of Alberta, Although not common, child-targeted and sports-related Edmonton, Alberta, Canada. Prowse et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:39 Page 11 of 11 Received: 27 November 2017 Accepted: 10 April 2018 24. Velazquez CE, Black JL, Ahmadi N. Food and beverage promotions in Vancouver schools: a study of the prevalence and characteristics of in- school advertising, messaging, and signage. Prev Med Rep. 2015;2:757–64. 25. Government of British Columbia. Guidelines for food and beverage sales in B.C. schools. Victoria: Province of British Columbia; 2013. References 26. Pettigrew S, Rosenberg M, Ferguson R, Houghton S, Wood L. Game on: do 1. Brennan LK, Brownson RC, Orleans CT. Childhood obesity policy research children absorb sports sponsorship messages? Public Health Nutr. 2013;16: and practice: evidence for policy and environmental strategies. Am J Prev 2197–204. Med. 2014;46:e1–16. 27. Smith M, Jenkin G, Signal L, McLean R. Consuming calories and creating 2. Government of Alberta. Active Alberta 2011–2021. http://culture.alberta. cavities: beverages NZ children associate with sport. Appetite. 2014;81:209–17. ca/recreation/active-alberta/pdf/Active-Alberta-Policy.pdf.Accessed 24 28. Randall Conrad and Associates, Roma M. ARPA’s infrastructure committee: Nov 2017. operations survey summary report. Edmonton: Alberta Recreation and Parks 3. Olstad DL, Downs SM, Raine KD, Berry TR, McCargar LJ. Improving children’s Association; 2006. nutrition environments: a survey of adoption and implementation of 29. Naylor PJ, Wekken SV, Trill D, Kirbyson A. Facilitating healthier food nutrition guidelines in recreational facilities. BMC Public Health. 2011;11:423. environments in public recreation facilities: results of a pilot project in 4. Naylor PJ, Bridgewater L, Purcell M, Ostry A, Wekken SV. Publically funded British Columbia, Canada. J Park Recreat Adm. 2010;28:37–58. recreation facilities: obesogenic environments for children and families? Int 30. Prowse R. Food marketing to children in Canada: a settings-based scoping J Environ Res Public Health. 2010;7:2208–21. review on exposure, power and impact. Health Promot Chronic Dis Prev 5. Nelson TF, Stovitz SD, Thomas M, Lavoi NM, Bauer KW, Neumark-Sztainer D. Can. 2017;37:274–92. Do youth sports prevent pediatric obesity? A systematic review and 31. Brownell KD, Warner KE. The perils of ignoring history: big tobacco played commentary. Curr Sports Med Rep. 2011;10:360–70. dirty and millions died. How similar is big food? Milbank Q. 2009;87:259–94. 6. Inoue Y, Berg BK, Chelladurai P. Spectator sport and population health: a 32. Kirk SFL, Penney TL, Freedhoff Y. Running away with the facts on food and scoping study. J Sport Manage. 2015;29:705–25. fatness. Public Health Nutr. 2010;13:147–8. 7. Kelly B, Baur LA, Bauman AE, King L, Chapman K, Smith BJ. “Food company 33. Folta SC, Goldberg JP, Economos C, Bell R, Meltzer R. Food advertising sponsors are kind, generous and cool”: (mis)conceptions of junior sports targeted at school-age children: a content analysis. J Nutr Educ Behav. players. Int J Behav Nutr Phys Act. 2011;8:95. 2006;38:244–8. 8. Castonguay J. Sugar and sports age differences in children’s responses to a high sugar cereal advertisement portraying physical activities. Comm Res. 2015; https://doi.org/10.1177/0093650215587357. 9. Kelly B, King L, Bauman AE, Baur LA, Macniven R, Chapman K, et al. Identifying important and feasible policies and actions for health at community sports clubs: a consensus-generating approach. J Sci Med Sport. 2014;17:61–6. 10. Government of British Columbia. Healthier choices in vending machines in BC public buildings. Victoria: Province of British Columbia; 2014. 11. Alberta Health and Wellness. The Alberta nutrition guidelines for children and youth. Edmonton: Government of Alberta; 2010. 12. Government of Nova Scotia. Healthy eating in recreation and sport settings guidelines. 2015. 13. World Health Organization. A framework for implementing the set of recommendations on the marketing of foods and non-alcoholic beverages to children. 2012. 14. Lee NR, Kotler P. Social marketing: influencing behaviors for good. Thousand Islands: Sage; 2011. 15. Carter M-A, Edwards R, Signal L, Hoek J. Availability and marketing of food and beverages to children through sports settings: a systematic review. Public Health Nutr. 2012;15:1373–9. 16. Wolfenden L, Kingsland M, Rowland BC, Dodds P, Gillham K, Yoong SL, et al. Improving availability, promotion and purchase of fruit and vegetable and non sugar-sweetened drink products at community sporting clubs: a randomised trial. Int J Behav Nutr Phys Act. 2015;12:35. 17. Olstad DL, Goonewardene LA, McCargar LJ, Raine KD. Choosing healthier foods in recreational sports settings: a mixed methods investigation of the impact of nudging and an economic incentive. Int J Behav Nutr Phys Act. 2014;11:6. 18. World Health Organization: Set of recommendations on the marketing of foods and non-alcoholic beverages to children. 2010. 19. Prowse RJL, Naylor P-J, Olstad DL, Carson V, Masse L, Storey K, et al. Reliability and validity of novel tool to comprehensively assess food and beverage marketing in recreational sport settings. Int J Behav Nutr Phys Act. 2018. https://doi.org/10.1186/s12966-018-0667-3. 20. Cohen J. Statistical power analysis for the behavioral sciences. revised ed. New York: Academic Press; 1977. 21. Carter MA, Signal L, Edwards R, Hoek J, Maher A. Food, fizzy, and football: promoting unhealthy food and beverages through sport - a New Zealand case study. BMC Public Health. 2013;13:126. 22. Kelly B, Baur LA, Bauman AE, King L, Chapman K, Smith BJ. Food and drink sponsorship of children’s sport in Australia: who pays? Health Promt Int. 2010;26:188–95. 23. Kelly B, Bauman AE, Baur LA. Population estimates of Australian children’s exposure to food and beverage sponsorship of sports clubs. J Sci Med Sport. 2014;17:394–8.

Journal

International Journal of Behavioral Nutrition and Physical ActivitySpringer Journals

Published: May 31, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off