Food advertising during children s television in Canada and the UK

Transcription

1 1 Institute of Health and Soiety, William Leeh Building, Newastle University, Newastle upon Tyne, UK; 2 Department of Family Relations and Applied Nutrition, University of Guelph, Guelph, Ontario, Canada; 3 Department of Community and Family Mediine, Duke University Shool of Mediine, Durham, North Carolina, USA Correspondene to: Jean Adams, Institute of Health and Soiety, William Leeh Building, Newastle University, Newastle upon Tyne NE2 4HH, UK; j.m.adams@nl.a.uk Aepted 13 January 2009 Published Online First 28 May 2009 Food advertising during hildren s television in Canada and the UK J Adams, 1 K Hennessy-Priest, 2 S Ingimarsdóttir, 1 J Sheeshka, 2 T Østbye, 3 M White 1 ABSTRACT Bakground: Television advertisements for less healthy foods are thought to ontribute to overweight and obesity in hildren. In the UK, new regulations on television food advertising to hildren ame into effet in April These prohibit advertisements for less healthy foods during or around programmes of partiular appeal to (OPAT) hildren. In Canada, self-regulated odes of pratie on television food advertising to hildren were reently strengthened. Objetive: To doument the nutritional ontent of food advertised and number of advertisements OPAT hildren broadast in the UK and entral Canada before the introdution of the new UK regulations. Design: advertisements broadast on four popular hannels in Canada and the three terrestrial ommerial hannels in the UK during 1 week in 2006 were identified and linked to relevant nutritional data. Food advertisements OPAT hildren and for less healthy produts were identified using the riteria in the UK regulations. Results: 2315 food related advertisements broadast in Canada and 1365 broadast in the UK were inluded % were for less healthy produts; 5 11% were OPAT hildren. Around 5% of food advertisements would have been prohibited under the new UK regulations. There were few differenes in the nutritional ontent of food desribed in advertisements that were and were not OPAT hildren. Conlusion: There was little evidene that food desribed in advertisements OPAT hildren were any less healthy than those that were not. Few food advertisements are likely to be prohibited by the new UK regulations. The inreasing prevalene of overweight and obesity in hildren in developed ountries 1 poses a signifiant threat to the long term health of those affeted. 2 A positive relationship between time spent wathing television (TV) and body weight has been onsistently doumented among hildren. 3 5 One possible reason for this is the large number of advertisements for less healthy foods on TV. 6 Exposure to TV food advertising has an independent influene on hildren s food preferenes, purhasing behaviour and purhasing requests. 6 There is the additional potential that TV advertising of less healthy foods normalises suh produts. 7 Conerns over TV food advertising to hildren have led to inreasing alls for regulation In the UK, Ofom, the ommuniations industry regulator, implemented new regulations in April Television advertisements on nonhildren s hannels for foods identified as less What is already known on this topi New regulations on television food advertising to hildren were implemented in the UK in 2007 Industry regulated odes of pratie were strengthened in Canada. The nutritional ontent of foods advertised on TV to hildren in the UK and Canada, and the effet of new regulations, is not known. What this study adds There were few differenes in the type or nutritional ontent of food advertisements that were and were not of partiular appeal to hildren. Foods advertisements of partiular appeal to hildren were no less healthy than those that were not. Only around 5% of food advertisements would have been prohibited by the new UK regulations. healthy 13 were prohibited during or around programmes of partiular appeal to (OPAT) 4 9 year. The prohibition was extended to programmes OPAT 4 15 year in January 2008 with a more phased introdution on hildren s hannels. Programmes OPAT hildren are defined as those where the proportion of individuals wathing who are hildren is at least 20% more than the proportion of hildren in the population. In Canada, less speifi self-regulated advertising odes of pratie on TV food advertising to hildren were reently strengthened. 14 Although there are likely to be international variations in TV food advertising, reent data suggest that TV food advertising in the UK and Canada is remarkably similar. 15 Numerous ontent analyses have doumented the types of food advertised on TV However, few authors have desribed the detailed nutritional ontent of the foods advertised Most previous analyses of food advertising to hildren have also foused speifially on hildren s programmes and hannels, wrongly assuming that hildren s viewing is restrited to these times. 25 Some notable exeptions that have studied advertising during the programmes that hildren wath most have been published in the last few years 23 but no reent data are available from the UK or Canada. Arh Dis Child: first published as /ad on 28 May Downloaded from on 27 Marh 2019 by guest. Proteted by opyright. 658 Arh Dis Child 2009;94: doi: /ad

2 Thus, there are urrently few detailed data available on the exat nutritional ontent of the foods that are being advertised on TV to hildren living in the UK or Canada. Furthermore, there are no published data on the number of advertisements, or on the nutritional ontent of the foods advertised, that will be subjet to the new Ofom regulations in the UK. In order to provide baseline data from before the introdution of the UK Ofom regulations and the Canadian odes of pratie, we ompared the type, nutritional ontent and proportion of food advertisements for less healthy produts amongst all TV food advertisements with those OPAT hildren in the UK and entral Canada. METHODS Data soures Information on all advertisements broadast on four popular free-to-view hannels in Ontario and Quebe, Canada (CBC (Toronto), CTV (Toronto), A-hannel and SRC (Montreal)) and all ommerial terrestrial hannels in the UK (ITV, Channel 4 and Channel 5) during the week beginning 30 Otober 2006 was obtained from audiene researh bureaux. These data inluded what produts were advertised and an indiator of the number of viewers of eah advertisement both overall and in age-speifi groups (termed television ratings (TVR) in the UK and gross rating points (GRP) in Canada). The data overed 24 h and were not limited to any partiular time of day. Identifying advertisements OPAT hildren TVR and GRP indiate the proportion of all individuals, or speifi demographi groups, that wathed any given advertisement. For the UK data, TVR were alulated based on viewing and population figures from aross the UK. For the Canadian data, GRP were alulated based on figures from Toronto, Ontario for the English language hannels (CBC (Toronto), CTV (Toronto) and A-hannel) and from Montreal, Quebe for the Frenh language hannel (SRC (Montreal)). For eah advertisement, TVR/GRP and population ount data were used to determine what proportion of individuals who were wathing eah advertisement were younger hildren and what proportion all hildren. Due to variations in the age groups for whih TVR/GRP data are available, younger hildren were defined as aged 4 9 years in the UK and 2 11 years in Canada, and all hildren as aged 4 15 years in the UK and 2 17 years in Canada. As defined in the Ofom regulations, those advertisements where the proportion of individuals wathing who were hildren was at least 20% more than the proportion of hildren in the referene population, were identified as OPAT hildren. Categorial analysis of food related advertisements related advertisements (inluding those for both food and beverages, food stores and restaurants, but exluding those for vitamins and other supplements) were identified and ategorised using a previously developed shema (table 1). 16 The proportion of food related advertisements that fell into eah ategory that were and were not OPAT younger and all hildren were ompared using Fisher s exat test. Nutritional analysis of food advertisements and estimation of the TV diet For food advertisements (ie, exluding advertisements for food stores and restaurants), information on energy, protein, arbohydrate, sugar, fat, alohol, fibre and sodium ontent and suggested portion size was obtained from pakaging and manufaturers websites as far as possible, supplemented with standard food table data in the UK 28 where neessary. For advertisements for brand ranges rather than speifi produts, a single default produt was identified based on the authors onsensus judgment of the most popular produts within brand ranges, and the nutritional ontent for that produt used in analyses. We summarised the nutritional ontent of foods advertised using the onept of the TV diet. 16 To determine the omposition of the TV diet, eah advertisement was assumed to ontribute one reommended sized portion of the produt advertised. The total perentage of energy from eah maronutrient, as well as fibre and sodium density (in grams per megajoule) for all foods advertised was then alulated. The omposition of the TV diet was alulated for all food advertisements and for those OPAT younger and all hildren separately, and the results in both the UK and Canadian data sets ompared. Finally, the UK Food Standards Ageny s nutrient profiling model 13 was used to identify whether or not advertised foods were less healthy. Differenes in the proportion of food advertisements that were less healthy, between those advertisements that were and were not OPAT younger and all hildren were assessed using Fisher s exat test. RESULTS Overall, 2315 food related advertisements broadast in Canada and 1365 broadast in the UK were inluded. In the Canadian sample, 105 (4.5%) food-related advertisements were OPAT younger hildren and 152 (6.6%) OPAT all hildren. Comparable figures from the UK were 139 (10.2%) and 144 (10.5%). The proportion of food advertisements that were OPAT younger and all hildren was signifiantly higher in the UK ompared to Canada (Fisher s exat test p,0.001 in both ases). The distributions of food related advertisements aross the food ategories are shown in table 1 along with the results of Fisher s exat tests omparing the frequeny of eah ategory in those advertisements that were and were not OPAT hildren. Few statistially signifiant differenes were seen. In Canada, advertisements for saues, stoks and ondiments were more ommon amongst advertisements OPAT younger hildren, and those for sweets and andy and food stores were less ommon amongst advertisements OPAT all hildren. In the UK, advertisements for grain produts were more ommon and those for food stores less ommon amongst advertisements OPAT both younger and all hildren. Advertisements for milk and milk produts were also more ommon amongst advertisements OPAT younger hildren in the UK. The nutritional ontent of food advertisements (the TV diet) is summarised in table 2. In both Canada and the UK there was a onsistent trend for food advertisements OPAT both younger and all hildren to have a markedly higher fibre density than all food advertisements. In Canada food advertisements OPAT younger and all hildren had a notieably higher alohol ontent than all advertisements. The reverse trend was seen in the UK. Food advertisements OPAT all hildren had a somewhat lower sugar ontent than all food advertisements in Canada. Table 3 shows the proportions of food advertisements that were for less healthy foods. Advertisements for less healthy foods aounted for more than half of all food advertisements in all ases but made up a higher proportion of food advertisements in Canada than in the UK (Fisher s exat test p,0.001). The proportion of food advertisements that were for less healthy foods did not differ signifiantly Arh Dis Child: first published as /ad on 28 May Downloaded from on 27 Marh 2019 by guest. Proteted by opyright. Arh Dis Child 2009;94: doi: /ad

3 Table 1 Food ategory Food and food related advertisements in Canada and UK by food ategory Canada, n (%) UK, n (%) 2 11 year between advertisements that were and were not OPAT younger or all hildren in either ountry. DISCUSSION This is the first detailed nutritional analysis of TV food advertising to hildren in either the UK or Canada that identified hildren s TV using viewing figures. Furthermore, this is the first published analysis of TV food advertising based on the new UK Ofom regulations on TV food advertising to hildren. Five months before the implementation of these regulations, only 10 11% of food advertisements in the UK were OPAT hildren and would be subjet to the regulations. In the Canadian data, 5 7% of food advertisements were OPAT hildren. There were few differenes in either ountry between the types, or nutritional ontent, of foods that were promoted during advertisements that were and were not OPAT hildren. In all ases, more than half of food advertisements were for less healthy produts. There was no evidene that the proportion of food advertisements that were for less healthy produts differed between food advertisements that were and were not OPAT hildren in either ountry. Overall, 5 6% of food advertisements in the UK sample would have been prohibited under the new regulations. We used industry data on what advertisements were broadast, who saw them and the nutritional ontent of foods advertised. Manufaturers data on nutrient omposition represent the most produt-speifi information available year 4 9 year 4 15 year Beverages 241 (10.4) 11 (10.5) 21 (13.8) 215 (15.8) 17 (12.2) 19 (13.2) Fisher s exat test p value 1.00* 0.17{ 0.26* 0.47{ Milk and milk produts 247 (10.7) 8 (7.6) 14 (9.2) 73 (5.4) 14 (10.1) 12 (8.3) Fisher s exat test p value 0.42* 0.68{ 0.02* 0.11{ Fruits, vegetables and juies 26 (1.1) 2 (1.9) 1 (0.7) 77 (5.6) 6 (4.3) 6 (4.2) Fisher s exat test p value 0.33* 1.00{ 0.57* 0.57{ Salty snaks 7 (0.3) Fisher s exat test p value 1.00* 1.00{ 2* 2{ Grain produts 249 (10.8) 12 (11.4) 20 (13.2) 182 (13.3) 30 (21.6) 32 (22.2) Fisher s exat test p value 0.75* 0.34{ 0.005* 0.002{ Meat and alternatives 24 (1.0) 1 (1.0) 2 (1.3) 15 (1.1) 1 (0.7) 1 (0.7) Fisher s exat test p value 1.00* 0 (0.95){ 1.00* 1.00{ Butter, margarine and spreads (1.3) 1 (0.7) 1 (0.7) Fisher s exat test p value 2* 2{ 1.00* 0.71{ Sweets and andy 211 (9.1) 7 (6.7) 6 (3.9) 97 (7.1) 6 (4.3) 9 (6.3) Fisher s exat test p value 0.49* 0.02{ 0.22* 0.86{ Chewing gum 11 (0.5) 2 (1.9) 1 (0.7) Fisher s exat test p value 0.09* 0.53{ 2* 2{ Saues, stoks and ondiments 43 (1.9) 7 (6.7) 6 (3.9) 99 (7.3) 12 (8.6) 12 (8.3) Fisher s exat test p value 0.003* 0.06{ 0.49* 0.61{ Meals 818 (35.3) 35 (33.3) 59 (38.8) 219 (16.0) 29 (20.9) 29 (20.1) Fisher s exat test p value 0.75* 0.38{ 0.11* 0.19{ Restaurants 329 (14.2) 19 (18.1) 20 (13.2) Fisher s exat test p value 0.25* 0.81{ 2* 2{ Food stores 109 (4.7) 1 (1.0) 2 (1.3) 370 (27.1) 23 (16.5) 23 (16.0) Fisher s exat test p value 0.06* 0.04{ 0.002* 0.001{ related adverts 2315 (100) 105 (100) 152 (100) 1365 (100) 139 (100) 144 (100) OPAT, of partiular appeal to. *Comparing advertisements of partiular appeal to younger hildren (2 11 or 4 9 year ) with advertisements not of partiular appeal to them using Fisher s exat test. {Comparing advertisements of partiular appeal to all hildren (2 17 or 4 15 year ) with advertisements not of partiular appeal to them using Fisher s exat test. Given the dependene of the advertising industry on audiene researh bureaux viewing figures, they are also likely to be aurate. The UK Ofom regulations on TV food advertising to hildren refer to advertisements during or around programmes OPAT hildren. In ontrast, we have foused on advertisements OPAT hildren. By fousing on advertisements shown during and around programmes OPAT hildren, the Ofom regulations assume that audienes for programmes are very similar to those for the advertisements shown during and around them. Early November was hosen as a typial period as being not too lose to either season premieres or the holiday season. However, seasonal variations in food advertising may exist. Similarly, variations may exist in food advertising aross parts of Canada and hannels not studied here. Our use of default produts to represent the nutrient ontent of brand ranges advertised may have introdued some error. Although a very similar pattern of results to those reported here was found when using the mean nutrient ontent of all produts in the brand ranges advertised (data not shown), the assumption that advertisements for brand ranges are omparable to advertisements for single produts may be wrong. As previously, we used standard portion sizes to determine the weight or volume of advertised produts that ontributed to the overall TV diet. While standard portion sizes may substantially underestimate the true weight of produts that individuals onsume, they appear to do so to a onsistent Arh Dis Child: first published as /ad on 28 May Downloaded from on 27 Marh 2019 by guest. Proteted by opyright. 660 Arh Dis Child 2009;94: doi: /ad

4 Table 2 Nutrient Nutritional ontent of foods advertised in Canada and the UK Canada ads (n = 1877) Food 2 11 year (n = 85) degree. 29 Our approah to quantifying the TV diet is, therefore, likely to represent a good estimation of the nutritional omposition of a diet omprised of one real-life portion of every food advertised. However, the onept of the TV diet is intended more as a method of summarising advertised produts than an indiation of what television viewers atually eat. Advertisements are likely to vary in their ability to enourage onsumption amongst viewers. As these analyses are desriptive only, they provide no information on the effet of advertisements on atual behaviour (although the advertisements raison d être is, of ourse, to influene behaviour). Finally, due to variations in data availability, the age ranges used to define younger and all hildren varied between the UK and Canada and omparisons between the UK and Canadian data should be interpreted autiously. Between 5% and 11% of food advertisements met the definition of OPAT hildren. We assume these represent the food advertisements that, in the UK, are now subjet to the Ofom regulations on TV food advertising to hildren. In the UK sample, 5 6% of food advertisements (52 61% of 10 11%) would have been prohibited by the new regulations. Given the apparent politial ommitment to these regulations, it is surprising that suh a small proportion of food advertisements will be affeted. Our results are likely to be heavily reliant on the definition of what makes an advertisement OPAT hildren that we used (ie, when the proportion of individuals wathing who were hildren was at least 20% more than the proportion of hildren in the referene population). This is the definition set out in the Ofom regulations and is, therefore, highly poliy relevant. However, other ut-offs or definitions may have resulted in different patterns of results. It is possible that when our data were olleted (5 months before the introdution of the Ofom regulations) food manufaturers advertising in the UK were already hanging their advertising strategies in antiipation of the regulations. However, the proportion of food advertisements that were Food 2 17 year (n = 130) UK ads (n = 995) Food 4 9 year (n = 116) Protein (% energy) Carbohydrate (% energy) Sugar (% energy) Fibre (g/mj) Fat (% energy) Alohol (% energy) Sodium (g/mj) OPAT, of partiular appeal to. Table 3 Healthfulness Food 4 15 year (n = 121) OPAT hildren was even lower in Canada than in the UK, suggesting that the low proportion of food advertisements that are OPAT hildren is not UK speifi. In addition, more than half of food advertisements OPAT hildren were for less healthy produts in both ountries and there was no evidene that the proportion of less healthy produts varied between advertisements that were and were not OPAT hildren in either ountry. Any hanges in TV food advertising to hildren in the UK in advane of the introdution in the Ofom regulations were not refleted in differenes between the UK and Canada at this time. Further work will be required to determine how the food advertising landsape in the UK hanges following full introdution of the Ofom regulations. Few differenes were seen between advertisements that were and were not OPAT hildren in either the type or nutrient ontent of foods, or the proportion of foods that were less healthy, in either ountry. This hallenges a ommon pereption that food advertisements broadast during hildren s TV are partiularly unhealthy ompared to advertisements broadast at other times. In fat, the one onsistent differene between advertisements that were and were not OPAT hildren was a higher fibre density in foods desribed in advertisements OPAT hildren. Two papers from the USA desribe the nutritional ontent of foods advertised during programmes popular with hildren. Although methods vary, two lear trends are apparent: the perentage of energy derived from sugar is substantially lower in our samples than previously (around 16% here ompared to 34% 24 and 46 49% 23 previously) and the perentage of energy derived from fat is muh higher in our samples than previously (around 35% 24 here ompared to 27% and 18 19% 23 previously). This onfirms that there are substantial international variations in the nutritional ontent of foods advertised on TV. CONCLUSIONS This is the first published study exploring TV food advertising to hildren in the ontext of the Ofom regulations implemen- Proportion of food advertisements that met the FSA definition of less healthy in Canada and the UK Canada, n (%) UK, n (%) 2 11 year 2 17 year 4 9 year 4 15 year Less healthy 1234 (65.7) 51 (60.0) 78 (60.0) 542 (54.5) 60 (51.7) 74 (61.2) Fisher s exat test p value 0.29* 0.15{ 0.55* 0.12{ FSA, UK Food Standards Ageny; OPAT, of partiular appeal to. *Comparing advertisements of partiular appeal to younger hildren (2 11 or 4 9 year ) with advertisements not of partiular appeal to them using Fisher s exat test. {Comparing advertisements of partiular appeal to all hildren (2 17 or 4 15 year ) with advertisements not of partiular appeal to them using Fisher s exat test. Arh Dis Child: first published as /ad on 28 May Downloaded from on 27 Marh 2019 by guest. Proteted by opyright. Arh Dis Child 2009;94: doi: /ad

5 ted in the UK in April Our data, olleted before the implementation of the new regulations, provide a useful baseline against whih future data an be ompared. We found little evidene of substantial differenes in the type, nutritional ontent or prevalene of less healthy foods between those advertisements that were and were not OPAT hildren in either the UK or Canada. Five months before the introdution of the Ofom regulations only around 5 6% of food advertisement on the three main UK ommerial hannels would have been prohibited by the regulations. Funding: JA is supported by a UK MRC Speial Training Fellowship in Health Servies and Health of the Publi Researh. Competing interests: None. REFERENCES 1. Speiser PW, Rudolf MC, Anhalt H, et al. Consensus statement: hildhood obesity. J Clin Endorinol Metab 2005;90: Reilly JJ, Methven E, MDowell ZC, et al. Health onsequenes of obesity. Arh Dis Child 2003;88: Robinson TN. Television viewing and hildhood obesity. Pediatr Clin North Am 2001;48: Caroli M, Argentieri L, Cardone M, et al. Role of television in hildhood obesity prevention. Int J Obes 2004;28:S Coon KA, Tuker KL. Television and hildren s onsumption patterns: a review of the literature. Minerva Pediatr 2002;54: Hastings G, Stead M, MDermott L, et al. Review of researh on the effets of food promotion to hildren, final report for the Food Standards Ageny. Glasgow: Centre for Soial Marketing, Anon. The elephant in the room: evolution, behaviouralism, and ounteradvertising in the oming war against obesity. Harv Law Rev 2003;116: Nestle M. Food marketing and hildhood obesity - matter of poliy. N Engl J Med 2006;354: Halford JCG. Servinguptrouble? Advertisingfoodto hildren. Psyhologist 2005;5: Department of Health. Choosing health: making healthy hoies easier. London: HMSO, Publi Health Assoiation of Australia In. Television food advertising during hildren s viewing times position statement. Deakin, ACT, Australia: Publi Health Assoiation of Australia In., Ofom. Television advertising of food and drink produts to hildren - final statement. London: Ofom, Rayner M, Sarborough P, Boxer A, et al. Nutrient profiles: development of final model. Final report for the Food Standards Ageny. Oxford: British Heart Foundation Health Promotion Researh Group, Department of Publi Health, University of Oxford, Canadian Children s Food and Beverage Advertising Initiative. Self-regulation of hildren s advertising in Canada. See hildrensinitiative/selfregulation.html (aessed 17 April 2009). 15. Adams J, Hennessy-Priest J, Ingimarsdóttir S, et al. Changes in food advertisements during prime-time television from 1991 to 2006: a UK-Canada omparison. Br J Nutr 2009 Feb 25:1 10. [Epub ahead of print] 16. Østbye T, Pomerleau J, White M, et al. Food and nutrition in Canadian prime time television ommerials. Can J Publi Health 1993;84: Chapman K, Niholas P, Supramaniam R. How muh food advertising is there on Australian television? Health Promot Int 2006;21: Connor SM. Food-related advertising on preshool television: building brand reognition in young viewers. Pediatris 2006;118: Lewis MK, Hill AJ. Food advertising on British hildren s television: a ontent analysis and experimental study with nine-year. Int J Obes Relat Metab Disord 1998;22: Rodd HD, Patel V. Content analysis of hildren s television advertising in relation to dental health. Br Dent J 2005;199: Chestnutt IG, Ashraf FJ. Television advertising of foodstuffs potentially detrimental to oral health a ontent analysis and omparison of hildren s and primetime broadasts. Community Dent Health 2002;19: Dibb S, Gordon S. TV dinners: what s being served up by the advertisers? London: Sustain: the alliane for better food and farming, Powell LM, Szzypka G, Chaloupka FJ, et al. Nutritional ontent of television food advertisements seen by hildren and adolesents in the United States. Pediatris 2007;120: Harrison K, Marske AL. Nutritional ontent of foods advertised during the television programs hildren wath most. Am J Publi Health 2005;95: Ofom. Childhood obesity - food advertising in ontext: hildren s food hoies, parents understanding and influene, and the role of food promotion. London: Ofom, nomis. Offiial labour market statistis. See (aessed 17 April 2009). 27. Statistis Canada. Table Estimates of population, by age group and sex for July 1, Canada, provines and territories, annual. CANSIM (database). Available at &Array_Pik=1&Detail=1&ResultTemplate=CII/CII &RootDir=CII/ (aessed 17 April 2009). 28. Food Standards Ageny. MCane and Widdowson s the omposition of foods. 6th edn. Cambridge: Royal Soiety of Chemistry, Welten DC, Carpenter RA, MPherson RS, et al. Comparison of a dietary reord using reported portion size versus standard portion size for assessing nutrient intake. Publi Health Nutr 2000;3: Butland B, Jebb S, Kopelman P, et al. Foresight takling obesities: future hoies - projet report. London: Government Offie for Siene, Arh Dis Child: first published as /ad on 28 May Downloaded from on 27 Marh 2019 by guest. Proteted by opyright. 662 Arh Dis Child 2009;94: doi: /ad