bs_bs_banner
doi:10.1111/jpc.12488
ORIGINAL ARTICLE
Industry self-regulation and TV advertising of foods to Australian children Lisa G Smithers, John W Lynch and Tracy Merlin School of Population Health, University of Adelaide, Adelaide, South Australia, Australia
Aim: The aim of this study is to examine the amount of non-core (unhealthy) food advertising currently on Australian television (i) during children’s programmes and viewing times; (ii) since the introduction of food industry self-regulatory initiatives in 2009; and (iii) whether advertising differs according to signatory status to industry initiatives. Methods: We systematically searched PubMed, Embase.com and JSTOR (media/marketing) databases; grey literature; and reference lists of relevant articles for studies published since 2009 that reported on food advertising on Australian television. Results: The title and abstract of 316 articles were screened, yielding 25 articles considered potentially eligible, of which eight met the pre-defined selection criteria. Meta-analysis was not possible because of temporal and methodological differences across studies. The advertising of non-core foods was found to be negligible during programmes with a C-(children’s) classification but ranged from 1.5 to 6.5/h during children’s peak viewing times. From 2006 to 2011, non-core food advertising decreased by 0.18 advertisements per hour every year, whereas fast food advertising increased by 0.09/h; however, these analyses are based on one study with only five time points. During children’s viewing times, signatories to industry initiatives advertise non-core foods at higher rates than non-signatories. Conclusions: Although it is not possible to determine whether advertising has changed since the industry initiatives were introduced, signatories to the initiatives continue to advertise non-core foods at times when many children watch television. Future efforts to reduce children’s exposure to food advertising should be focused on advertising during children’s peak viewing times rather than by programme classifications. Key words:
advertising; child; food and beverage; food industry; television.
What is already known on this topic
What this paper adds
1 The advertising of non-core foods on television is linked to poorer diets in children. 2 The food industry has introduced voluntary codes to address food marketing to Australian children. 3 Television is a major source of exposure to food advertising among school children.
1 There is a negligible food advertising during programmes carrying the C (Children’s) classification; however, Australian children are exposed to non-core foods that are advertised when they watch programmes with other classifications. 2 Non-core foods are advertised on Australian television in higher proportions than healthy foods. 3 Although it is unclear whether the voluntary initiatives have changed food advertising behaviour, signatories to the initiatives currently advertise non-core foods on Australian television at higher rates than non-signatories.
Introduction The marketing of non-core foods is thought to contribute to the obesogenic environment. Reducing children’s exposure to non-core food advertising is one of the Australian National Preventive Health Taskforce’s recommendations for preventing obesity.1 Large-scale systematic reviews conducted in the USA, Correspondence: Dr Lisa G Smithers, School of Population Health, Mail drop DX 650550, University of Adelaide, Adelaide, SA 5005, Australia. Fax: +61 8 8313 3511; email: [email protected] Conflict of interest: The authors have no conflicts of interest to declare. Accepted for publication 21 October 2013.
386
UK and by the World Health Organisation conclude that food advertising influences children’s preferences for foods, their food purchases or requests to care givers for purchases, and ultimately the foods children consume.2–4 To address community concern about the advertising of noncore foods to children, industry groups in Australia introduced two voluntary self-regulation initiatives. In January 2009, the Australian Food and Grocery Council introduced the responsible children’s marketing initiative (RCMI), and in August 2009, the quick service restaurant industry (QSRI) in conjunction with the Australian Association of National Advertisers introduced a similar initiative for the fast food industry. Signatories to these initiatives commit to not advertise food products
Journal of Paediatrics and Child Health 50 (2014) 386–392 © 2013 The Authors Journal of Paediatrics and Child Health © 2013 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
LG Smithers et al.
to children unless the products represent healthy dietary choices (as per established scientific or Australian government standards), and the advertising message is in the context of a healthy life-style that encourages good dietary and physical activity habits. Seventeen food companies are signatories to the RCMI, and seven are QSRI signatories. There has been no systematic examination of evidence to determine whether these initiatives have influenced the advertising of non-core foods. The aims of this review were as follows: (i) to examine the amount of non-core food advertising currently on Australian television during children’s peak viewing times (PVTs) and according to programme classifications (C ‘children’, P ‘pre-school’ and other); (ii) to examine whether the frequency of non-core food advertising to children has changed since the RCMI/QSRI were introduced; and (iii) to compare signatories and non-signatories’ advertising of noncore foods during children’s programmes. The focus of this review was specifically on television advertising because, on average, Australian children watch 2 h of television/day, and therefore, television remains a major source of exposure to food advertising.5,6 This review was commissioned by the Australian National Preventive Health Agency (ANPHA), and a comprehensive report is available on the ANPHA website (http:// www.anpha.gov.au).
Materials and Methods The review protocol was written by the authors and approved by ANPHA prior to commencement. No changes were made to the protocol as a result of ANPHA’s involvement.
Search strategy Systematic searches of Medline (PubMed), EMBASE.com and JSTOR were conducted using search terms tailored to each database and pilot tested (available on request). JSTOR was searched because it is a marketing and media database. Searches were limited to studies on humans and published in English since 2009. The decision to limit the search from 2009 was made a priori and based on ascertaining children’s current exposure to non-core food advertising and changes since the industry initiatives. Other search strategies included checking reference lists of eligible articles, using Google Scholar and PubMed to locate articles citing the papers included in the review and examining industry websites. Also, representatives from ANPHA contacted and invited researchers working in the field to provide unpublished articles. The last database search was conducted on 24 August 2012.
Eligibility All studies that examined television advertising to Australian children that were published since 2009 were eligible. According to the Participants, Intervention, Comparator, Outcome (PICO) criteria,7 participants were advertisers of non-core foods (subgroups by signatory status). Intervention refers to industry self-regulation initiatives (the RCMI/QSRI), comparator was no industry initiatives and outcomes were advertising of non-core foods during children’s programmes, PVT and advertising trends over time in absolute numbers, percentages and rates. The dif-
TV advertising of food to children
ferent definitions of non-core foods, PVT and children’s programmes used in eligible studies were accepted. No attempt was made to re-examine the evidence according to pre-defined standard definitions.
Study selection Titles and abstracts of articles were screened, and the full article was retrieved for potentially eligible studies. Several unpublished studies were obtained. Two articles that were in press were included, whereas another two articles that had not been through peer review were not included because of confidentiality concerns.
Data collection process Data were extracted systematically from each study using a purpose-built form. Data were extracted by LGS and verified by JWL and TM. Authors were contacted to clarify published data and to provide supplementary information.
Assessment of study quality and bias The evidence hierarchy described by Merlin et al.8 and recommended by National Health and Medical Research Council (NHMRC)9 was applied to assess the risk of bias within individual studies, and the execution of each study was critically appraised using an adapted Cochrane checklist.10 It was not possible to assess publication bias because of the small number of studies.
Synthesis of evidence Study findings were summarised using the FORM system for grading evidence.9,11 Although the protocol allowed scope for a meta-analysis, it was deemed inappropriate because of differences in the outcomes, methods and quality across studies.
Results The evidence base Figure 1 shows that eight of the 316 title/abstracts obtained from the literature search were included in the review. Table 1 provides a summary of the included studies.12–19 Where three articles were published using different data from the same study,14,17,18 the primary analysis was undertaken on the most recent paper that utilised all available data.17 Five studies reported non-core food advertising on free-to-air (FTA) television in major cities, one in a regional area and two involved subscription television (PayTV). Advertising data were manually recorded17 and/or sourced from media monitoring agencies.12–19 Children’s PVT was defined differently between studies but was commonly determined from audience viewing data or, for PayTV, from the channels most popular with children. The number of viewing days in each study ranged from 4 to 92. Categorisation of food advertisements as ‘core’ or ‘non-core’ was predominantly based on the Australian Guide to Health Eating (AGHE). However, the classification used in one industry report differed by signatory status such that AGHE or the NSW healthy school canteens strategy was applied to advertising by non-signatories, whereas for signatories, the company’s own nutrition criteria were used.12
Journal of Paediatrics and Child Health 50 (2014) 386–392 © 2013 The Authors Journal of Paediatrics and Child Health © 2013 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
387
TV advertising of food to children
LG Smithers et al.
programmes with G-(general) or other classifications (1.3 and 1.5, respectively).16 In industry reports, children’s programmes were defined according to the programme content or the audience share (>50% children), rather than using standard programme classifications. Industry reports suggest that 4.2% of all food advertisements were shown during children’s programmes and 2.4% were for non-core foods;13 thus, 57% of food advertisements shown during children’s programmes are for noncore foods (2.4/4.2 = 57%). Another industry report describes the number of unique advertisements shown during children’s programmes but not repetition,12 which is not comparable with other studies. Roberts et al. obtained broadcasters’ data to examine advertising during periods scheduled for children’s programmes and found 330 breaches of the RCMI/QSRI.19
Aim 2: changes in non-core food advertising following the introduction of the RCMI/QSRI Fig. 1 Flow of articles through the different stages of the systematic review.
Study quality Across all studies, advertising data were measured objectively so the RCMI/QSRI was unlikely to have influenced data collection. There were no missing data, and studies appeared free of selective reporting of outcomes. However, it is not clear whether the introduction of the RCMI/QSRI was independent of other secular changes to advertising, such as economic events that might impact on company advertising budgets. Trends in food advertising prior to the initiatives were not reported, and there was no examination of whether advertising is seasonal. Rationales were not provided for the dates or number of days sampled and, thus, whether the study was sufficiently powered to detect changes in advertising over time. The main results are presented according to each of the aims described in the introduction.
Aim 1: current advertising of non-core foods during children’s PVT and programmes Four studies reported non-core food advertising during children’s PVT13,15–17 and four during children’s programmes.12,13,16,19 Two studies that included data from multiple time points reported that the rate of non-core food advertising during children’s PVT varied between 3.1 and 5.9 advertisements per hour on metropolitan FTA stations,16,17 which is higher than the rate of 1.5/h from an industry report.13 For PayTV, non-core food advertising was consistent across two studies at 0.7–1.1/h, with rates varying widely according to channel. Some PayTV channels had no food advertisements (Nick Junior), whereas others advertised 2.8 non-core food advertisements/h (Fox8).15,16 The percentage of all food advertisements for non-core foods was stable at 50–60% on metropolitan FTA stations,16,17 slightly higher on regional FTA (70%)16 and increased on PayTV from 56% in March 2010 to 93% in November 2011.15,16 During C-classified programmes, non-core food advertisements ranged between 0 and 0.04/h, which is far lower than 388
Two studies reported non-core food advertising before and after the introduction of the industry initiatives (Fig. 2). Brindal et al. found that food advertising data sourced from different suppliers differed widely, and consequently, they were unable to examine trends over time.16 The reason behind this disparity remains unclear. King et al. showed that overall, food advertisements have declined by 0.23/h (95% confidence interval −0.38, −0.07) for every year between 2006 and 2011.17 When separated into subcategories, fast food advertising increased by 0.09/h (0.02, 0.15), whereas other non-core foods declined by 0.18/h (−0.30, −0.06).17 The study is limited in that there are fewer than three data points prior to the introduction of the industry initiatives.10 Other publications that report weaker pre- versus post-RCMI/QSRI analyses (level IV evidence) are not presented here.14,18
Aim 3: comparison of non-core food advertising by signatory status Four studies compared non-core food advertising according to signatory status.12,16,17,19 Upon request, Brindal et al. supplied data showing that since October 2009, signatories to the RCMI/ QSRI advertised non-core foods during children’s PVT at higher rates than non-signatories.16 For example, in November 2011, non-core food advertisements during children’s PVT were 0.36/h for RCMI signatories and 0.17/h for non-signatories, whereas QSRI signatories and non-signatories advertised at 0.21 and 0.05/h, respectively. In contrast, studies with fewer data points found that RCMI signatories advertised non-core foods during children’s PVT at a lower rate than non-signatories (1.5 vs. 2.0/h in 2009).14 Hebden et al. suggest that non-signatories to the QSRI had a larger reduction in fast food advertising from 2009 to 2010 than signatories,18 although additional data supplied by the authors indicate that this may be because of an increase in healthier fast food advertising from 0/h in 2009 to 0.3/h in 2010. Generally, a higher percentage of all food advertisements by RCMI/QSRI signatories than non-signatories occur during children’s PVT16,17 and children’s programmes.12 In November 2011, 70% of all advertisements by RCMI signatories during children’s PVT were for non-core foods compared with 22% for non-signatories.16
Journal of Paediatrics and Child Health 50 (2014) 386–392 © 2013 The Authors Journal of Paediatrics and Child Health © 2013 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
Design (Level of evidence)†
Post-test case series (Level IV)
Hebden (2011)15 Commercial monitors
Commercial monitors
Commercial monitors
Source of data
Journal of Paediatrics and Child Health 50 (2014) 386–392 © 2013 The Authors Journal of Paediatrics and Child Health © 2013 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
Post-test case series (Level IV)
Roberts (2012)19
T1, T2: manual collection T3 , T 4 , T 5 : Commercial monitors Ebiquity Adelaide Brisbane Perth Melbourne Sydney
Sydney
FTA
FTA
FTA & PayTV‡
PayTV
FTA
FTA
FTA PayTV
T1 Oct 08 T2 Feb 09‡ T3 May 09 T4 Oct 09 T5 Mar 10‡ T6 Jul 10‡ T7 Nov 11‡ T1 May 06 T2 May 07 T3 May 09 T4 Apr 10 T5 May 11 Sept–Oct 10
Feb 09
Mar 10
Mar–May 10
Dates
61 d 24 h/d
4–7 d 8–11 h/d
4d 16 h/d
4d 14 h/d
14 d 24 h/d
92 d 24 h/d
Period sampled
PVT: Weekdays: 0600–0900 + 1600–2100 Weekends: 0600–1200 + 1600–2100 Programme: Children’s programmes according to the scheduling for C-rated programmes that broadcasters supplied to ACMA
PVT: if >35% of child audience (PayTV >75%)§ Weekdays: 1730–2200 Weekends: 0730–1100 + 1630–2200
Programmes: if >50% of audience was 50% audience was
doi:10.1111/jpc.12488
ORIGINAL ARTICLE
Industry self-regulation and TV advertising of foods to Australian children Lisa G Smithers, John W Lynch and Tracy Merlin School of Population Health, University of Adelaide, Adelaide, South Australia, Australia
Aim: The aim of this study is to examine the amount of non-core (unhealthy) food advertising currently on Australian television (i) during children’s programmes and viewing times; (ii) since the introduction of food industry self-regulatory initiatives in 2009; and (iii) whether advertising differs according to signatory status to industry initiatives. Methods: We systematically searched PubMed, Embase.com and JSTOR (media/marketing) databases; grey literature; and reference lists of relevant articles for studies published since 2009 that reported on food advertising on Australian television. Results: The title and abstract of 316 articles were screened, yielding 25 articles considered potentially eligible, of which eight met the pre-defined selection criteria. Meta-analysis was not possible because of temporal and methodological differences across studies. The advertising of non-core foods was found to be negligible during programmes with a C-(children’s) classification but ranged from 1.5 to 6.5/h during children’s peak viewing times. From 2006 to 2011, non-core food advertising decreased by 0.18 advertisements per hour every year, whereas fast food advertising increased by 0.09/h; however, these analyses are based on one study with only five time points. During children’s viewing times, signatories to industry initiatives advertise non-core foods at higher rates than non-signatories. Conclusions: Although it is not possible to determine whether advertising has changed since the industry initiatives were introduced, signatories to the initiatives continue to advertise non-core foods at times when many children watch television. Future efforts to reduce children’s exposure to food advertising should be focused on advertising during children’s peak viewing times rather than by programme classifications. Key words:
advertising; child; food and beverage; food industry; television.
What is already known on this topic
What this paper adds
1 The advertising of non-core foods on television is linked to poorer diets in children. 2 The food industry has introduced voluntary codes to address food marketing to Australian children. 3 Television is a major source of exposure to food advertising among school children.
1 There is a negligible food advertising during programmes carrying the C (Children’s) classification; however, Australian children are exposed to non-core foods that are advertised when they watch programmes with other classifications. 2 Non-core foods are advertised on Australian television in higher proportions than healthy foods. 3 Although it is unclear whether the voluntary initiatives have changed food advertising behaviour, signatories to the initiatives currently advertise non-core foods on Australian television at higher rates than non-signatories.
Introduction The marketing of non-core foods is thought to contribute to the obesogenic environment. Reducing children’s exposure to non-core food advertising is one of the Australian National Preventive Health Taskforce’s recommendations for preventing obesity.1 Large-scale systematic reviews conducted in the USA, Correspondence: Dr Lisa G Smithers, School of Population Health, Mail drop DX 650550, University of Adelaide, Adelaide, SA 5005, Australia. Fax: +61 8 8313 3511; email: [email protected] Conflict of interest: The authors have no conflicts of interest to declare. Accepted for publication 21 October 2013.
386
UK and by the World Health Organisation conclude that food advertising influences children’s preferences for foods, their food purchases or requests to care givers for purchases, and ultimately the foods children consume.2–4 To address community concern about the advertising of noncore foods to children, industry groups in Australia introduced two voluntary self-regulation initiatives. In January 2009, the Australian Food and Grocery Council introduced the responsible children’s marketing initiative (RCMI), and in August 2009, the quick service restaurant industry (QSRI) in conjunction with the Australian Association of National Advertisers introduced a similar initiative for the fast food industry. Signatories to these initiatives commit to not advertise food products
Journal of Paediatrics and Child Health 50 (2014) 386–392 © 2013 The Authors Journal of Paediatrics and Child Health © 2013 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
LG Smithers et al.
to children unless the products represent healthy dietary choices (as per established scientific or Australian government standards), and the advertising message is in the context of a healthy life-style that encourages good dietary and physical activity habits. Seventeen food companies are signatories to the RCMI, and seven are QSRI signatories. There has been no systematic examination of evidence to determine whether these initiatives have influenced the advertising of non-core foods. The aims of this review were as follows: (i) to examine the amount of non-core food advertising currently on Australian television during children’s peak viewing times (PVTs) and according to programme classifications (C ‘children’, P ‘pre-school’ and other); (ii) to examine whether the frequency of non-core food advertising to children has changed since the RCMI/QSRI were introduced; and (iii) to compare signatories and non-signatories’ advertising of noncore foods during children’s programmes. The focus of this review was specifically on television advertising because, on average, Australian children watch 2 h of television/day, and therefore, television remains a major source of exposure to food advertising.5,6 This review was commissioned by the Australian National Preventive Health Agency (ANPHA), and a comprehensive report is available on the ANPHA website (http:// www.anpha.gov.au).
Materials and Methods The review protocol was written by the authors and approved by ANPHA prior to commencement. No changes were made to the protocol as a result of ANPHA’s involvement.
Search strategy Systematic searches of Medline (PubMed), EMBASE.com and JSTOR were conducted using search terms tailored to each database and pilot tested (available on request). JSTOR was searched because it is a marketing and media database. Searches were limited to studies on humans and published in English since 2009. The decision to limit the search from 2009 was made a priori and based on ascertaining children’s current exposure to non-core food advertising and changes since the industry initiatives. Other search strategies included checking reference lists of eligible articles, using Google Scholar and PubMed to locate articles citing the papers included in the review and examining industry websites. Also, representatives from ANPHA contacted and invited researchers working in the field to provide unpublished articles. The last database search was conducted on 24 August 2012.
Eligibility All studies that examined television advertising to Australian children that were published since 2009 were eligible. According to the Participants, Intervention, Comparator, Outcome (PICO) criteria,7 participants were advertisers of non-core foods (subgroups by signatory status). Intervention refers to industry self-regulation initiatives (the RCMI/QSRI), comparator was no industry initiatives and outcomes were advertising of non-core foods during children’s programmes, PVT and advertising trends over time in absolute numbers, percentages and rates. The dif-
TV advertising of food to children
ferent definitions of non-core foods, PVT and children’s programmes used in eligible studies were accepted. No attempt was made to re-examine the evidence according to pre-defined standard definitions.
Study selection Titles and abstracts of articles were screened, and the full article was retrieved for potentially eligible studies. Several unpublished studies were obtained. Two articles that were in press were included, whereas another two articles that had not been through peer review were not included because of confidentiality concerns.
Data collection process Data were extracted systematically from each study using a purpose-built form. Data were extracted by LGS and verified by JWL and TM. Authors were contacted to clarify published data and to provide supplementary information.
Assessment of study quality and bias The evidence hierarchy described by Merlin et al.8 and recommended by National Health and Medical Research Council (NHMRC)9 was applied to assess the risk of bias within individual studies, and the execution of each study was critically appraised using an adapted Cochrane checklist.10 It was not possible to assess publication bias because of the small number of studies.
Synthesis of evidence Study findings were summarised using the FORM system for grading evidence.9,11 Although the protocol allowed scope for a meta-analysis, it was deemed inappropriate because of differences in the outcomes, methods and quality across studies.
Results The evidence base Figure 1 shows that eight of the 316 title/abstracts obtained from the literature search were included in the review. Table 1 provides a summary of the included studies.12–19 Where three articles were published using different data from the same study,14,17,18 the primary analysis was undertaken on the most recent paper that utilised all available data.17 Five studies reported non-core food advertising on free-to-air (FTA) television in major cities, one in a regional area and two involved subscription television (PayTV). Advertising data were manually recorded17 and/or sourced from media monitoring agencies.12–19 Children’s PVT was defined differently between studies but was commonly determined from audience viewing data or, for PayTV, from the channels most popular with children. The number of viewing days in each study ranged from 4 to 92. Categorisation of food advertisements as ‘core’ or ‘non-core’ was predominantly based on the Australian Guide to Health Eating (AGHE). However, the classification used in one industry report differed by signatory status such that AGHE or the NSW healthy school canteens strategy was applied to advertising by non-signatories, whereas for signatories, the company’s own nutrition criteria were used.12
Journal of Paediatrics and Child Health 50 (2014) 386–392 © 2013 The Authors Journal of Paediatrics and Child Health © 2013 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
387
TV advertising of food to children
LG Smithers et al.
programmes with G-(general) or other classifications (1.3 and 1.5, respectively).16 In industry reports, children’s programmes were defined according to the programme content or the audience share (>50% children), rather than using standard programme classifications. Industry reports suggest that 4.2% of all food advertisements were shown during children’s programmes and 2.4% were for non-core foods;13 thus, 57% of food advertisements shown during children’s programmes are for noncore foods (2.4/4.2 = 57%). Another industry report describes the number of unique advertisements shown during children’s programmes but not repetition,12 which is not comparable with other studies. Roberts et al. obtained broadcasters’ data to examine advertising during periods scheduled for children’s programmes and found 330 breaches of the RCMI/QSRI.19
Aim 2: changes in non-core food advertising following the introduction of the RCMI/QSRI Fig. 1 Flow of articles through the different stages of the systematic review.
Study quality Across all studies, advertising data were measured objectively so the RCMI/QSRI was unlikely to have influenced data collection. There were no missing data, and studies appeared free of selective reporting of outcomes. However, it is not clear whether the introduction of the RCMI/QSRI was independent of other secular changes to advertising, such as economic events that might impact on company advertising budgets. Trends in food advertising prior to the initiatives were not reported, and there was no examination of whether advertising is seasonal. Rationales were not provided for the dates or number of days sampled and, thus, whether the study was sufficiently powered to detect changes in advertising over time. The main results are presented according to each of the aims described in the introduction.
Aim 1: current advertising of non-core foods during children’s PVT and programmes Four studies reported non-core food advertising during children’s PVT13,15–17 and four during children’s programmes.12,13,16,19 Two studies that included data from multiple time points reported that the rate of non-core food advertising during children’s PVT varied between 3.1 and 5.9 advertisements per hour on metropolitan FTA stations,16,17 which is higher than the rate of 1.5/h from an industry report.13 For PayTV, non-core food advertising was consistent across two studies at 0.7–1.1/h, with rates varying widely according to channel. Some PayTV channels had no food advertisements (Nick Junior), whereas others advertised 2.8 non-core food advertisements/h (Fox8).15,16 The percentage of all food advertisements for non-core foods was stable at 50–60% on metropolitan FTA stations,16,17 slightly higher on regional FTA (70%)16 and increased on PayTV from 56% in March 2010 to 93% in November 2011.15,16 During C-classified programmes, non-core food advertisements ranged between 0 and 0.04/h, which is far lower than 388
Two studies reported non-core food advertising before and after the introduction of the industry initiatives (Fig. 2). Brindal et al. found that food advertising data sourced from different suppliers differed widely, and consequently, they were unable to examine trends over time.16 The reason behind this disparity remains unclear. King et al. showed that overall, food advertisements have declined by 0.23/h (95% confidence interval −0.38, −0.07) for every year between 2006 and 2011.17 When separated into subcategories, fast food advertising increased by 0.09/h (0.02, 0.15), whereas other non-core foods declined by 0.18/h (−0.30, −0.06).17 The study is limited in that there are fewer than three data points prior to the introduction of the industry initiatives.10 Other publications that report weaker pre- versus post-RCMI/QSRI analyses (level IV evidence) are not presented here.14,18
Aim 3: comparison of non-core food advertising by signatory status Four studies compared non-core food advertising according to signatory status.12,16,17,19 Upon request, Brindal et al. supplied data showing that since October 2009, signatories to the RCMI/ QSRI advertised non-core foods during children’s PVT at higher rates than non-signatories.16 For example, in November 2011, non-core food advertisements during children’s PVT were 0.36/h for RCMI signatories and 0.17/h for non-signatories, whereas QSRI signatories and non-signatories advertised at 0.21 and 0.05/h, respectively. In contrast, studies with fewer data points found that RCMI signatories advertised non-core foods during children’s PVT at a lower rate than non-signatories (1.5 vs. 2.0/h in 2009).14 Hebden et al. suggest that non-signatories to the QSRI had a larger reduction in fast food advertising from 2009 to 2010 than signatories,18 although additional data supplied by the authors indicate that this may be because of an increase in healthier fast food advertising from 0/h in 2009 to 0.3/h in 2010. Generally, a higher percentage of all food advertisements by RCMI/QSRI signatories than non-signatories occur during children’s PVT16,17 and children’s programmes.12 In November 2011, 70% of all advertisements by RCMI signatories during children’s PVT were for non-core foods compared with 22% for non-signatories.16
Journal of Paediatrics and Child Health 50 (2014) 386–392 © 2013 The Authors Journal of Paediatrics and Child Health © 2013 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
Design (Level of evidence)†
Post-test case series (Level IV)
Hebden (2011)15 Commercial monitors
Commercial monitors
Commercial monitors
Source of data
Journal of Paediatrics and Child Health 50 (2014) 386–392 © 2013 The Authors Journal of Paediatrics and Child Health © 2013 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
Post-test case series (Level IV)
Roberts (2012)19
T1, T2: manual collection T3 , T 4 , T 5 : Commercial monitors Ebiquity Adelaide Brisbane Perth Melbourne Sydney
Sydney
FTA
FTA
FTA & PayTV‡
PayTV
FTA
FTA
FTA PayTV
T1 Oct 08 T2 Feb 09‡ T3 May 09 T4 Oct 09 T5 Mar 10‡ T6 Jul 10‡ T7 Nov 11‡ T1 May 06 T2 May 07 T3 May 09 T4 Apr 10 T5 May 11 Sept–Oct 10
Feb 09
Mar 10
Mar–May 10
Dates
61 d 24 h/d
4–7 d 8–11 h/d
4d 16 h/d
4d 14 h/d
14 d 24 h/d
92 d 24 h/d
Period sampled
PVT: Weekdays: 0600–0900 + 1600–2100 Weekends: 0600–1200 + 1600–2100 Programme: Children’s programmes according to the scheduling for C-rated programmes that broadcasters supplied to ACMA
PVT: if >35% of child audience (PayTV >75%)§ Weekdays: 1730–2200 Weekends: 0730–1100 + 1630–2200
Programmes: if >50% of audience was 50% audience was
