Please be advised that due to the high volume of submissions, the ANZCTR is currently experiencing delays in processing submissions from those outside of Australia and New Zealand. As the ANZCTR is funded by Australia and New Zealand, we must prioritise submissions from these countries first, therefore international submissions should allow additional time for registration or updates to registration records. Apologies for any inconvenience caused.

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
ACTRN12617001230347
Ethics application status
Approved
Date submitted
18/08/2017
Date registered
23/08/2017
Date last updated
23/08/2017
Type of registration
Retrospectively registered

Titles & IDs
Public title
Investigation into the sustained impact of food advertising on children’s dietary intake and its potential to influence children’s weight
Scientific title
Investigation into the sustained impact of energy–dense food advertising on children’s dietary intake: a within–subject, randomised, crossover, counter–balanced trial
Secondary ID [1] 292678 0
Nil known
Universal Trial Number (UTN)
U1111-1200-9155
Trial acronym
N/A
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Childhood overweight and obesity 304430 0
Unhealthy diet 304436 0
Condition category
Condition code
Diet and Nutrition 303760 303760 0 0
Obesity
Public Health 303777 303777 0 0
Other public health

Intervention/exposure
Study type
Interventional
Description of intervention(s) / exposure
Brief summary:
The aim of this research was to investigate the direct effects of children’s exposure to food advertising on their dietary intake and its potential influence in the development of childhood overweight. Across a series of six-day holiday camps, children (n=160, aged 7-12 years) were exposed to food and non-food advertising in an online game and/or a television cartoon in a randomised within-subject, cross-over, counterbalanced design. The primary aim of the study is to explore whether short-term increases in snack intake following food advertising exposure are compensated for at a subsequent lunchtime meal; and hence identify if food advertising contributes to a positive energy balance likely to contribute to childhood overweight.

Brief background:
There is a recognised link between unhealthy food marketing and childhood obesity yet major regulatory reform remains essentially unimplemented. Children continue to be exposed to high levels of unhealthy food marketing across a wide range of media and settings. A key reason for the failure to restrict marketing to children has been the shortage of evidence showing a direct link between food marketing and children’s weight. The aspect of food marketing that will be the focus of this study is food promotion, specifically television advertising and online advertising (in the form of branded games (‘advergames’) ) and their potential impact on children’s food consumption.

Short term experimental studies, designed to assess the effect of unhealthy food advertising (embedded in either a cartoon or in a game) and children’s subsequent food intake, show there is an immediate direct effect from single exposure (children consume more snack foods). This study aims to observe what happens to children when they are exposed to food advertising over a longer period; to investigate the effect of food advertising across multiple media; and to explore the potential impact of exposure on longer term dietary intake.

Setting:
We partnered with the University of Wollongong (UOW) School Holiday Sports Camp and UOW’s Early Start to conduct this study. Early Start is a child-focused research facility on the same campus as the holiday camp. The study was conducted as part of usual care at the UOW holiday camp during the morning session (8am-1.30pm). Camps run during each school holiday period, four times a year. Morning session camp fees for children participating in the study were fully funded. The interventions were conducted in rooms within Early Start. A pilot study was run during the school holiday period in January 2016, followed by the main four studies which were run during the school holidays in April, July and September 2016 and January 2017, each for a period of six days.

Study population:
Children aged 7-12 years (n=160 (n=40 x 4 camps))

Study design:
The study utilised a within–subject, randomised, crossover, counterbalanced design with two media condition arms: multiple media (TV plus advergame) or single media (TV only). Each media arm had two advertising conditions: control (an advergame featuring a non–food brand and/or exposure to ten non–food TV advertisements) and experimental (an advergame featuring a food brand and/or exposure to ten TV food advertisements).

Two groups of 20 children were formed for each camp, with an approximate even spread of gender and age between groups. The two groups were then randomised to one of the two media arms (single or multiple media). Within each holiday camp children remained within the media arm to which they were allocated, and completed both the food advertising and non-food advertising conditions within their allocated media group. Each advertising condition lasted for three days with the order of advertising exposure counterbalanced across holiday camps. The food advertising condition was first in the April and September camps and second in the July and January camps. The April, September and January camps had a washout period of four days between Days 1-3 and Days 4-6. There was no washout period during the July camp: this was constrained due to the camp organisers only running the children’s holiday camp over a six day period.

Materials:
The TV advertisements (approximately 30 seconds each) were embedded within a ten minute age–appropriate, gender–neutral cartoon and shown each day. The cartoons were matched by series across condition but with a different episode of the cartoon shown each day. There were no references to, or depictions of, foods or eating in any of the cartoons screened. In order to isolate the effects of the study advertising exposure, the branded products selected for the experimental condition were for real products available in other countries but not available for sale in Australian supermarkets or advertised on commercial TV within Australia.

Daily procedure:
Each morning at 8am, children arrived fasting (self-reported by parents/caregivers) at the Early Start research centre and signed in by their parents/caregivers, who then left. Children washed their hands and were then served breakfast. The dining space was set up with 40 individual trays and, each day, children were offered a selection of portioned breakfast cereals, fruit, toast or pikelets, spreads and water to drink. Four research assistants supervised the children. The children were told that they could eat as much or as little as they liked, and were given more of each food item as requested. Children were allocated a unique identifying number which was placed on their meal tray at each eating occasion and used to track how much each child ate daily. After 30 minutes, the children left to participate in the nearby holiday camp activities. Camp leaders ensured that the physical intensity of camp activities was similar each session and day of the sports camp. Children’s breakfast intake was quantified by assessing whether children had eaten all, more than half, half, less than half or none of each food item. Kilojoules (kJ) were estimated from the proportion of each standardised food item consumed, and children’s mean breakfast intakes for the three days of food and three days of non–food advertising were estimated.

At mid–morning children were directed to their intervention rooms in the research centre with each room supervised by two trained research assistants. All children washed their hands and a roll call was taken. The rooms were bright and colourful with a carpeted floor and a large, wall mounted TV screen. Children in both intervention rooms were asked to take a seat on the floor and to report how hungry they felt by filling out a picture rating scale. Children were then told that they would be watching a cartoon and some advertisements. Once the cartoon had finished, iPads were distributed to the children in the multiple media condition and, after an explanation on how to play, they played an advergame for 5 minutes. In both rooms, after media exposure was complete, individual trays with six small bowls of snack foods were given out. Bowls contained 50 grams of each of the following foods: high fat savoury (crinkle–cut crisps, plain potato crisps or chicken–flavoured crackers); low fat savoury (pretzels, plain crackers or rice crackers); high fat sweet (milk chocolate, chocolate–covered biscuits or sugar–coated chocolate confectionery); low fat sweet (assorted jelly lollies); fruit (green and black grapes or peeled mandarin segments); and vegetable (carrot sticks). These selections were in line with previous study designs. Snack items offered on Day One were matched with those offered on Day Four, Day Two with Day Five and Day Three with Day Six. None of the brands offered to children to eat were featured in any of the advertisements. Children were asked to wait until everyone had their trays before they could start eating. They were, again, told that they could eat as little or as much as they liked, and if they would like some more they should put their hand up and they would be given more of the requested food. This was the only time either food or eating was referred to by the research assistants. Advertisements were not discussed with the children. Eating time was limited to 15 minutes after which children were instructed to stand up and leave the room. Afterwards children left to take part in further holiday camp activities and the remaining food on each child’s tray was weighed and recorded.

Lunch was served at 13.00 each day back in the research centre dining room. Once again 40 individual trays were set up with pre–weighed food items. There was a different menu for Days One to Three which was repeated on Days Four to Six. Lunch items included fruit, vegetables, yoghurt and healthier versions of fast food, e.g. low fat beef burger, oven baked chicken pieces and chips. The dining room was supervised by a minimum of 4 research assistants. All children washed their hands before eating. Each day prior to lunch, children completed the picture rating scale to report how hungry they felt. As with the previous meal occasions, children were told that they could eat as little or as much as they liked and if they would like some more to ask and they would be given more of the requested food. Parents arrived to pick children up at 13.30, whereupon children were signed out. Once children had left, all remaining individual food items were weighed and recorded. We did not collect dietary data from children once they left camp for the day.

Parents and children completed online questionnaires at baseline (prior to study commencement) and 1 week post study completion. Parent questionnaires included child and family baseline characteristics (including Birch et al's (2001) validated Child Feeding Questionnaire). Other questionnaire content is detailed in the "Outcomes" section.
Intervention code [1] 298913 0
Behaviour
Comparator / control treatment
The trial is a randomised crossover-trial, each participant will be exposed to food advertising (experimental condition) and non-food advertising (control condition). Children's food intake and behavioural responses after each condition will be compared with each other.
Control group
Active

Outcomes
Primary outcome [1] 303124 0
Children's morning snack intake (kJ)
Each day all snack food items were weighed (g) prior to being presented to participants and then reweighed after the eating period was finished to calculate how much of each food the children had eaten. This was conducted by a trained research assistant (qualified dietitian) in the absence of the children. Children’s morning snack intakes were converted from gram amounts to kilojoules using FoodWorks 8 nutrient analysis software. Mean snack food intakes (kJ) were calculated for all children for the three days of food advertising and three days of non-food advertising.
Timepoint [1] 303124 0
Each day of the study after advertising exposure (6 days)
Primary outcome [2] 303125 0
Children's lunch food intake (kJ)
Each day all lunch food items were weighed (g) prior to being presented to participants and then reweighed after the eating period was finished. This was conducted by a trained research assistant (qualified dietitian) in the absence of the children. Children’s lunch intakes were converted from gram amounts to kilojoules using FoodWorks 8 nutrient analysis software.
Timepoint [2] 303125 0
Each day of the study after the lunchtime period (6 days)
Secondary outcome [1] 337949 0
Children's daily breakfast intake (kJ)
Children's breakfast intake was estimated from the proportion of each standardised food item consumed and their mean breakfast intakes for the three days of food and three days of non–food advertising were estimated. This was conducted by a trained research assistant (qualified dietitian) in the absence of the children.
Timepoint [1] 337949 0
Each day of the study after the breakfast period (6 days)
Secondary outcome [2] 337951 0
Children’s self-reported hunger prior to morning tea and lunch
Children reported how hungry they felt prior to morning tea and lunch each day using a validated picture–rating scale; with the anchors “I am really hungry” and “I am not hungry at all”. Picture rating scales are straight forward to use for children of this age group and are a reliable tool for assessing children’s subjective experiences related to hunger and satiety.
Timepoint [2] 337951 0
Each day of the study prior to advertising exposure and prior to lunchtime
Secondary outcome [3] 337953 0
Children’s height and weight.
These were measured on Day 1 of the study by two trained research assistants. Children’s body mass index (BMI) were calculated and these values were used to classify children into underweight, normal weight, overweight or obese categories using international standardised cut–points. These measurements were taken in a quiet corner of the room by a trained research assistant. Neither the child, nor any subsequent child being measured, saw the measurements recorded. If any children was reluctant to be weighed or measured they were not coerced to do so.
Timepoint [3] 337953 0
Day 1 of the study.
Secondary outcome [4] 337954 0
Children’s recognition of advertised foods.
An online brand recognition tool was designed to assess children’s recollection of the advertised food brands, both pre and post study. Children were asked: a) if they recognised 20 different photographs of both food and non–food logos, and b) to describe the product to which the logo related. If children reported recognising a particular logo or brand, they were asked a series of questions designed to assess children’s attitudes and associations to/with that particular brand. This tool was based on a validated food brand recognition instrument for children of this age group. Children completed the online brand recognition questionnaires at home prior to the study commencing and in the week following the study’s completion and a brand recognition score was calculated.
Timepoint [4] 337954 0
At baseline and one week post-study
Secondary outcome [5] 337960 0
Children's comments about advertisements and purchase requests
In an online questionnaire, designed for this study, parents were asked:
1. Whether their child had made any comments or talked about the advertisements they had seen during the study
2. To report any purchase requests their child might have made for any of the advertised snack food products shown during the study period. They were asked to report the name of the product and how many times the children had asked for it.
This questionnaire was completed one week after the study had finished.
Timepoint [5] 337960 0
1 week post study

Eligibility
Key inclusion criteria
Children aged 7 to 12 years were eligible to participate if they were able to attend all six days of the holiday camp. Informed, written ethical consents were obtained from all children’s parents or guardians.
Minimum age
7 Years
Maximum age
12 Years
Gender
Both males and females
Can healthy volunteers participate?
Yes
Key exclusion criteria
Children were considered ineligible if they reported having any food allergies or intolerances; any medical conditions affecting what they could eat (eg diabetes); had any dietary restrictions (eg vegetarian); reported a dislike of the study foods; were not able to sit still and focus on a task for at least 15 minutes

Study design
Purpose of the study
Prevention
Allocation to intervention
Randomised controlled trial
Procedure for enrolling a subject and allocating the treatment (allocation concealment procedures)
Allocation to media condition was concealed. After recruitment was complete for each study camp (ie 40 children were recruited), two groups of 20 children were formed, with an approximate even spread of gender and age between groups. These groups were then randomly allocated to a media group by an independent researcher.
Methods used to generate the sequence in which subjects will be randomised (sequence generation)
A simple, manual randomisation method was used, with the first group of 20 children drawn out of a hat allocated to the single media intervention. This was conducted by an independent researcher not associated with the study.
Masking / blinding
Open (masking not used)
Who is / are masked / blinded?



Intervention assignment
Crossover
Other design features
Within-subject design. Order of advertising exposure was counterbalanced by camp to minimise any order effects (April and September camps: food advertising first 3 days; July and January camps: food advertising last three days). Participants were not made aware of the aim of the study.
Phase
Not Applicable
Type of endpoint(s)
Efficacy
Statistical methods / analysis
The sample size, with sufficient statistical power (80%) to assess the first primary outcome, with a significance of 0.05, was estimated from published data from a similar, short–term advertising exposure feeding trial in the UK using the differences in kilojoules reported between advertising conditions.

Analysis of the primary outcomes will be conducted using linear mixed models, adjusting for the clustered nature of the data (i.e. camp identifier will be included as a random intercept in the models). The linear mixed models will be used to examine the differences in snack intake (kJ) between the two media groups and the differences in the snack and lunch intakes (kJ) between advertising conditions within each group. Any influence of the impact of children's age (months), gender, weight status (BMI z–score), children’s baseline brand recognition score, household weekly income, hunger, parental feeding practice on snack intake will be investigated by adding these variables as covariates to the model. In addition, linear mixed models will be used to examine any differences in the number of purchase requests and brand recognition and attitudes between the two media conditions. All analyses will use a significance level of 0.05.

All analyses will be completed using the Statistical Package for the Social Sciences statistical software package, version 23 (SPSS Inc., Chicago, IL, USA).

Recruitment
Recruitment status
Completed
Date of first participant enrolment
Anticipated
Actual
Date of last participant enrolment
Anticipated
Actual
Date of last data collection
Anticipated
Actual
Sample size
Target
Accrual to date
Final
Recruitment in Australia
Recruitment state(s)
NSW

Funding & Sponsors
Funding source category [1] 297318 0
Government body
Name [1] 297318 0
Australian Research Council
Address [1] 297318 0
Australian Research Council,
Level 2, 11 Lancaster Place,
Canberra Airport,
ACT 2609 AUSTRALIA
Country [1] 297318 0
Australia
Funding source category [2] 297324 0
Charities/Societies/Foundations
Name [2] 297324 0
Cancer Council NSW
Address [2] 297324 0
153 Dowling St,
Woolloomooloo
NSW 2011
AUSTRALIA
Country [2] 297324 0
Australia
Primary sponsor type
Individual
Name
Bridget Kelly
Address
University of Wollongong,
School of Health and Society,
Northfields Avenue,
Wollongong, NSW 2522
AUSTRALIA
Country
Australia
Secondary sponsor category [1] 296289 0
None
Name [1] 296289 0
N/A
Address [1] 296289 0
N/A
Country [1] 296289 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 298420 0
University of Wollongong Human Research Ethics Committee
Ethics committee address [1] 298420 0
Research Services Office,
University of Wollongong,
Northfields Avenue,
Wollongong,
NSW 2522
AUSTRALIA
Ethics committee country [1] 298420 0
Australia
Date submitted for ethics approval [1] 298420 0
29/09/2015
Approval date [1] 298420 0
30/10/2015
Ethics approval number [1] 298420 0
HE15/396

Summary
Brief summary
The aim of this research was to investigate the direct effects of children’s exposure to food advertising on their dietary intake and its potential influence on the development of childhood overweight. Across a series of six-day holiday camps, children (n=160, aged 7-12 years) were exposed to food and non-food advertising in an online game and/or a television cartoon in a randomised within-subject, cross-over, counterbalanced design. The primary aim of the study is to explore whether short-term increases in snack intake following food advertising exposure are compensated for at a subsequent lunchtime meal; and hence identify if food advertising contributes to a positive energy balance likely to contribute to childhood overweight.
Trial website
Trial related presentations / publications
Norman J, Kelly B, McMahon A-T, Boyland E, Baur L, Bauman A, King L, Chapman K and Hughes C (2017). The sustained impact of unhealthy food advertising on children’s dietary intake: results from an experimental study. Paper presented at the International Society of Behavioral Nutrition and Physical Activity Conference, Victoria, BC, Canada, 8th June 2017
Public notes
Attachments [1] 1979 1979 0 0
Attachments [2] 1980 1980 0 0

Contacts
Principal investigator
Name 77030 0
Dr Bridget Kelly
Address 77030 0
Early Start
School of Health and Society
University of Wollongong
Northfields Avenue
Wollongong 2522
NSW
Country 77030 0
Australia
Phone 77030 0
+61 2 4221 3893
Fax 77030 0
Email 77030 0
bkelly@uow.edu.au
Contact person for public queries
Name 77031 0
Mrs Jennifer Norman
Address 77031 0
Early Start
School of Health and Society
University of Wollongong
Northfields Avenue
Wollongong 2522
NSW
Country 77031 0
Australia
Phone 77031 0
+61 407998874
Fax 77031 0
Email 77031 0
jan20@uowmail.edu.au
Contact person for scientific queries
Name 77032 0
Mrs Jennifer Norman
Address 77032 0
Early Start
School of Health and Society
University of Wollongong
Northfields Avenue
Wollongong 2522
NSW
Country 77032 0
Australia
Phone 77032 0
+61 407998874
Fax 77032 0
Email 77032 0
jan20@uowmail.edu.au

No information has been provided regarding IPD availability
Summary results
Have study results been published in a peer-reviewed journal?
Other publications
Have study results been made publicly available in another format?
Results – basic reporting
Results – plain English summary