Purpose: High levels of screen time (ST) can be detrimental to adolescent health. However, many Australian adolescents are currently exceeding national ST guidelines (≤2h/day). Public health concerns may rise, especially now that technology is increasingly part of the adolescent landscape. However, there are few longitudinal studies on ST trends among Australian adolescents. Therefore, the purpose of this study was to examine longitudinal trends in total and domain-specific ST between 2010 and 2014 in Australian adolescents.

Methods: Data were derived from the Longitudinal Study of Australian Children (LSAC). In 2010, 2,179 children (49.7% boys; 10.3 ± 1.1 years) completed a time-use diary, recording their main activities during waking hours. This was repeated with the same sample in 2012 (12.4 ± 0.5 years) and 2014 (14.4 ± 0.5 years). Data were analyzed for time spent in several screen-based activities, including TV-viewing, computer use (excluding games), electronic gaming, social networking and online communication. Repeated-measures ANCOVAs were performed to analyze trends in ST, controlling for sex, household income, maturational status, and waist circumference. Trends in ST were also analyzed by sex.

Results: Total ST significantly increased (+85.9min/day) between 2010 and 2014 (P < .001), and the prevalence of adolescents exceeding ST guidelines (>2h/day) increased markedly (2010; 59.5% - 2014; 75.3%), but differed by sex, with a larger increase in boys (2010-2012; +41.6 vs +22.7min/day). Electronic gaming increased in boys (2010-2014; +43.2min/day), and decreased in girls (−16.8min/day). In contrast, girls reported larger increases in TV-viewing (2012-2014; boys: +0.4, girls: +29.1min/day), computer use (excluding games) (2010-2012; boys: +24.8, girls: +34.3min/day) and time communicating online and social networking (2010-2012; boys: +4.3, girls: +15.2min/day).

Conclusions: The amount of time adolescents spend on screens increases as they age, although this differs by sex and ST domain. Findings suggest that future ST reduction interventions should mainly focus on computer use and electronic gaming in boys and TV-viewing and time spent communicating online and social networking for girls. 

Purpose: Active video games (AVGs) can increase physical activity (PA). Animated narrative videos possess unique immersive and motivational properties and have been found to increase PA via AVG play compared to no video condition. However, there is no evidence of whether that was due to narratives or simply adding an animated video. We investigated the effects of an animated narrative video (NV) vs. an animated non-narrative video (N-NV) on moderate-to-vigorous PA (MVPA) levels during AVG play. 

Methods: We randomly assigned 22 children 8-12 years old without previous AVG experience, after consent/assent, anamnesis, and anthropometric assessment, to watch either an NV or N-NV (~11 minutes), after which we instructed them to play an AVG for as long as they wanted. We estimated the time spent (minutes) in MVPA levels from vector magnitude (VM) counts recorded during the AVG sessions with a hip worn-accelerometer. Afterwards, they reported their experience in narrative immersion via a questionnaire. We applied independent two-sample t-tests to compare groups, with Cohen’s d to determine the effect size. We used Pearson correlation to test for associations.   

Results: The NV and N-NV groups did not differ significantly (age: 9.45±1.13 vs. 10.30±1.34, p=0.14; BMI%: 41.95±24.97 vs. 57.23±36.16, p=0.28). The NV group had significantly higher narrative immersion (Cronbach’s α=0.62, 3.50±0.55 vs. 2.91±0.59, p=0.03, d=1.03) and significantly higher PA levels than the N-NV group (VM: 91.21±34.98 vs. 64.24±13.71, p=0.03, d=1.02; MVPA: 20.11±13.75 vs. 7.85±5.83; p=0.02; d=1.13). MVPA was positively correlated with narrative immersion (r=0.21; p=0.01).

Conclusions: We are the first to test the effect of a narrative vs. nonnarrative video to motivate children to engage in MVPA via AVG play. The addition of animated narrative videos resulted in higher narrative immersion and PA levels when compared to animated nonnarrative videos.

Purpose: The widespread use of online food ordering systems presents tremendous opportunities for public health nutrition. For example, online school canteen ordering systems are increasingly popular in Australian schools. The leading provider processes over 13 million lunch orders per year, placed by over 250,000 users, across more than 1,200 schools. This research program investigates the potential for a public health nutrition intervention embedded in this online system to improve the healthiness of student purchases. Specifically, the aim was to determine the short-term effectiveness of the ‘Click & Crunch’ intervention in improving the nutritional i) quality and ii) content of primary school students’ online lunches orders.

Methods: 17 non-government schools were recruited to this Cluster RCT. Schools randomised to the intervention received changes to the layout and presentation of their online menu (including menu labelling, item repositioning, prompts, tailored feedback, and incentives) as well as a brief menu feedback report. The application of the intervention strategies (e.g. menu labelling) was based on the classifications stipulated by the NSW Healthy School Canteen Strategy (‘Everyday’, ‘Occasional’). Control schools received no change to their online menu. Intervention effectiveness was determined by comparing the i) the average proportion of ‘Everyday’ menu items, and ii) the nutritional content of students’ online canteen orders (the average energy (kJ), saturated fat (g), sugar (g) and sodium (mg) content), using purchasing data that is automatically collected and stored by the online ordering system. Short-term follow up occurred in the school term immediately following the switching on of the intervention strategies.

Results: During each 10-week baseline and follow-up period, over 3,000 orders were placed. Preliminary analysis indicates that the proportion of ‘Everyday’ foods purchased increased by 5% and there were decreases in the kilojoule, saturated fat, sodium and sugar content of intervention students’ online lunch orders relative to controls, but that these differences were not significant (p>0.05).

Conclusions: There are challenges when embedding public health interventions within real-world online systems. However, the wide-reach of these systems facilitating efficient dissemination at scale warrant further investigation into how to successfully embed behaviour change strategies.

Purpose

Early adolescence is an important period for (mental) health promotion. mHealth interventions offer great opportunities; but disadvantages are the small response rate and low engagement to such interventions, especially among vulnerable groups. Studies show that the inclusion of social support increases participants’ engagement. This study aims to explore whether a chatbot is useful for that purpose. This presentation describes the iterative, participatory development of a chatbot complementing a self-regulation app (MOV-E-STAR) to motivate adolescents (12-15y) to adopt a healthy lifestyle.

Methods

The development of the MOV-E-STAR chatbot follows the five steps of the mHealth Development and Evaluation Framework: 1) conceptualization; 2) formative research; 3) pre testing; 4) pilot testing; and, 5) qualitative research for intervention refinement. In the formative stage, relevant messages (n=319) were retrieved from a free and anonymous online helpline for children and adolescents in Flanders (Belgium) to examine how adolescents ask questions about the included health domains. In addition, six focus groups were conducted to identify preferences for the input and design of the chatbot. Data were analysed using thematic analysis in Nvivo. After inserting all input into Dialogflow, the chatbot was pre-tested by a small group of adolescents (n=6).

A prototype of the chatbot will be pilot tested, followed by a process evaluation with interviews. Results of the pilot study will be available at the time of the conference.

Results

Adolescents in the focus groups expressed that they preferred a colourful Messenger/Whatsapp design, that the chatbot should have a human side, that answers from the chatbot should not be too long and must be formulated in a positive way, that answers could be supplemented with emojis, that referrals to websites should immediately lead to clear information, and that the chatbot should have options to personalize.

Conclusions

Developing an mHealth intervention is an iterative process, in which repetitive testing with the target group and stakeholders is required. Further development will be informed by adolescents’ feedback on their experience of the prototype intervention. Based upon their input, the intervention will be further optimised.

Purpose: Mobile health interventions are increasingly utilized to support change in adolescents’ health behaviours as this age group embraces technologies. However, there is need to develop innovative strategies to support engagement with such technologies. This study evaluated the extent to which Aim2Be version 1 supports adolescents and their parents to adopt healthy behaviours (dietary, physical activity (PA), sedentary behaviours) and whether the app influences the targeted mediators. Methods: 632 families were invited to use Aim2Be for 4.5 months - a lifestyle behaviour modification app targeted at adolescents and their parents. Aim2Be aligns with current health recommendations; integrates theories and behavior change techniques known to support behaviour change (self-regulation from Self-Determination Theory and social environment from Social Cognitive Theory); uses gamified elements to promote engagement; adds social discussion and live health coach, and links living green with behaviour change. Assessments included adolescents’ and parents’ self-reports of PA, fruit and vegetable (F&V) consumption, and screen time as well as adolescents’ mediators of behaviour change (self-efficacy and intrinsic motivation). Linear regressions, conducted in Stata (v.15), were used to assess changes. Relevant covariates were added (age, gender, socio-economic status, and total time in app measured with web-analytics). Results: 294 and 312 Canadian adolescents (14.8yrs, 49% boys) and parents (46.3yrs, 66% mothers, 68% Caucasian), respectively utilized the Aim2Be app. Overall, changes in behaviours or mediators were more apparent when time in the app was accounted for. Adolescents who used the app (≥30 minutes) significantly increased F&V intake, reduced consumption of 100% fruit juices, and decreased screen time as well as increasing their intrinsic motivation and self-efficacy related to dietary behaviours. Parents who used the app (≥30 minutes) significantly increased F&V intake and decreased consumption of sugar-sweetened beverages. Discussion/conclusions: Evaluation of Aim2Be provided some initial support for the app in impacting some aspects of adolescents’ and their parents’ health behaviours and the mediators targeted. These initial findings were used to fine tune the Aim2Be app in preparation for the randomized control trial (e.g., integrating self-monitoring tools and increasing emphasis on PA).

Purpose: Poor eating behaviours are common among Australians and are often established in childhood and have a high change of persisting throughout adolescence and into adulthood. Nutrition knowledge interventions have been found to be effective in changing eating behaviour. Primary school children are a popular population for early nutrition education. The use of computer games for the intent of nutrition education has previously been successful in changing health-related behaviours and teaching nutrition. The current study investigates the development and effects of a tablet-based educational game for improvement of nutrition knowledge.

Methods: ‘VitaVillage’ is a farming-style game, where the user converts a virtual village to a health-promoting village by completing quests posed by villagers in need of healthy foods and advice. The user needs to grow the foods on their farm by correctly answering questions about healthy foods and their nutritional value. Year 5/6 primary school students were recruited into either a control group (mean age: 10.9, SD 0.8, 53.2% female), or an intervention group (mean age: 10.9, SD 0.7, 51.5% female). Both groups completed a nutrition knowledge survey at baseline (T0), and one week later (T1). The control group received no nutrition education, while the intervention group played the VitaVillage game twice for 20 minutes each time.

Results: Results indicated there was a significant increase in total nutrition knowledge scores compared to the control group (ΔM2.3, p<0.05, Cohen’s d=0.35). On a likeability survey after the study, children reported that they liked the game overall, with mean score of 77 (SD 24.6), on a scale of 0-100. The comments made by the children were overall very positive and will assist in the ongoing development of VitaVillage.

Conclusion: The ability of VitaVillage’s efficacy to increase children’s nutrition knowledge is promising and the children enjoyed playing the game. Further development and a larger school based randomised controlled including a study arm with traditional nutrition education and measures of dietary intake and longer game exposure is needed to assess efficacy and for consideration of implementation into primary schools in the future.