You are here

  • Home
  • Archive
  • Volume 12, Issue 3
  • Protocol for the Let’s Grow randomised controlled trial: examining efficacy, cost-effectiveness and scalability of a m-Health intervention for movement behaviours in toddlers

Article Text

Article menu
Paediatrics
Protocol
Protocol for the Let’s Grow randomised controlled trial: examining efficacy, cost-effectiveness and scalability of a m-Health intervention for movement behaviours in toddlers
  1. Kylie D Hesketh1,
  2. Katherine L Downing1,
  3. Barbara C Galland2,
  4. Jan M Nicholson3,
  5. Rachael Taylor4,
  6. Liliana Orellana5,
  7. Mohamed Abdelrazek6,
  8. Harriet Koorts1,
  9. Victoria Brown7,
  10. Jess Haines8,
  11. Karen J Campbell1,
  12. Lisa M Barnett1,9,
  13. Marie Löf1,10,11,
  14. Marj Moodie7,
  15. Valerie Carson12,
  16. Jo Salmon1
  1. 1Institute for Physical Activity and Nutrition, Deakin University Faculty of Health, Geelong, Victoria, Australia
  2. 2Department of Women's and Children's Health, University of Otago, Dunedin, New Zealand
  3. 3Judith Lumley Centre, La Trobe University, Bundoora, Victoria, Australia
  4. 4Department of Medicine, University of Otago, Dunedin, New Zealand
  5. 5Faculty of Health Biostatistics Unit, Deakin University, Geelong, Victoria, Australia
  6. 6School of Information Technology, Deakin University, Burwood, Victoria, Australia
  7. 7Deakin Health Economics, Institute for Health Transformation, Deakin University Faculty of Health, Geelong, Victoria, Australia
  8. 8Department of Family Relations and Applied Nutrition, University of Guelph, Guelph, Ontario, Canada
  9. 9School of Health & Social Development, Deakin University Faculty of Health, Geelong, Victoria, Australia
  10. 10Department of Health, Medicine and Caring Sciences, Linköping University, Linkoping, Sweden
  11. 11Department of Biosciences and Nutrition, Karolinska Institutet, Stockholm, Sweden
  12. 12Faculty of Kinesiology, Sport, and Recreation, University of Alberta, Edmonton, Alberta, Canada
  1. Correspondence to Dr Kylie D Hesketh; kylie.hesketh{at}deakin.edu.au

Abstract

Introduction Despite being an important period for the development of movement behaviours (physical activity, sedentary behaviour and sleep), few interventions commencing prior to preschool have been trialled. The primary aim of this trial is to assess the 12-month efficacy of the Let’s Grow mHealth intervention, designed to improve the composition of movement behaviours in children from 2 years of age. Let’s Grow is novel in considering composition of movement behaviours as the primary outcome, using non-linear dynamical approaches for intervention delivery, and incorporating planning for real-world implementation and scale-up from its inception.

Methods and analysis A randomised controlled trial will test the effects of the 12-month parental support mHealth intervention, Let’s Grow, compared with a control group that will receive usual care plus electronic newsletters on unrelated topics for cohort retention. Let’s Grow will be delivered via a purpose-designed mobile web application with linked SMS notifications. Intervention content includes general and movement-behaviour specific parenting advice and incorporates established behaviour change techniques. Intervention adherence will be monitored by app usage data. Data will be collected from participants using 24-hour monitoring of movement behaviours and parent report at baseline (T0), mid-intervention (T1; 6 months post baseline), at intervention conclusion (T2; 12 months post baseline) and 1-year post intervention (T3; 2 years post baseline). The trial aims to recruit 1100 families from across Australia during 2021. In addition to assessment of efficacy, an economic evaluation and prospective scalability evaluation will be conducted.

Ethics and dissemination The study was approved by the Deakin University Human Ethics Committee (2020-077). Study findings will be disseminated through publication in peer-reviewed journals, presentation at scientific and professional conferences, and via social and traditional media.

Trial registration number ACTRN12620001280998; U1111-1252-0599.

  • Community child health
  • PREVENTIVE MEDICINE
  • PUBLIC HEALTH
http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

http://dx.doi.org/10.1136/bmjopen-2021-057521

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Strengths and limitations of this study

    • This mHealth trial targets improvement in all movement behaviours with assessment of a single primary compositional outcome.

    • Efficacy, economic and scale-up evaluations will each be conducted.

    • A key strength is incorporation of stakeholder involvement and prospective evaluation and planning for implementation and scale-up at the efficacy testing stage.

    Introduction

    Life-course studies suggest interventions in early life, when children are undergoing rapid development, provide a window of opportunity to alter trajectories and have sustained effects on health.1 Early childhood (0–5 years) is a key time to focus on crucial health behaviours that impact children’s immediate and later outcomes.2–4 From a young age, physical activity is associated with better motor skill development, fitness, cognitive development, cardiometabolic health, bone, skeletal and psychosocial health.2 Longer sleep duration is positively associated with healthy growth and psychosocial health, lower adiposity and lower injury risk.3 5 Sedentary behaviours (sitting or lying when awake with energy expenditure ≤1.5 metabolic equivalent) are unfavourably associated with adiposity, motor skill development, cognitive development and psychosocial health.4 Yet only 9% of Australian 20 month olds6 and 15% of 4 year olds7 achieve nationally8 and internationally recommended9 levels of all three movement behaviours (physical activity, sedentary behaviour and sleep). This is not a problem unique to Australia, similar adherence is observed internationally.10–12 Thus, considerable potential exists to improve adherence to all three guidelines.

    Although most intervention research has tended to consider these behaviours separately, it is increasingly recognised that a single integrated movement continuum exists,6 from sleep (no conscious movement) through to high intensity physical activity. The composition of a day is understood as the proportion of time spent in each of the three movement behaviours where the proportion of time spent in one behaviour is considered relative to the other two. In practical terms, with only 24 hours in a day, the more time a child spends in one movement behaviour, the less time they have available for another. This provides enhanced opportunities to promote behaviour change by targeting both direct change, through impacting the target behaviour, and indirect change, by targeting other movement behaviours (ie, substitution).

    Few early childhood health behaviour interventions have commenced prior to the preschool period (age 3–5 years), with the toddler period (age 1–2 years) particularly neglected.13 Only one intervention including toddlers (but with a broad age range for commencement extending from 2 to 10 years old) has aimed to optimise all three movement behaviours.14 It focused on community-based obesity prevention, also targeting diet, and showed impact on only one of the three movement behaviours, increased sleep duration.14 15 Another study commencing in infancy and extending into the toddler period targeted all three movement behaviours as well as diet, observing little improvement in any movement behaviour with intervention.16 Studies to date have conceptualised and analysed each movement behaviour individually. No studies have presented their findings using a compositional approach that takes into account the constrained nature of the data (the fact that there can only ever be 24 hours in a day). Further, the potential for future scale-up (delivery at scale within existing health systems) has not previously been assessed prospectively in early childhood behaviour interventions. Typically, hybrid effectiveness and implementation trials follow initial effectiveness trials,17 although even this is rare.

    Planning for real-world implementation and scale-up is essential for effective translation. For the purposes of this trial, we use the following definition of scalability: ‘the ability of a health intervention shown to be efficacious on a small scale and/or under controlled conditions to be expanded under real world conditions to reach a greater proportion of the eligible population while retaining effectiveness’.18 When efficacious interventions are scaled up and implemented in real-world settings, they can report lower effect sizes19 and may be less likely to be sustained over time.20 21 Increasing research into evaluations of effective interventions that are capable of sustainable practice translation is a public health priority.22 23 A major challenge to practice translation is that intervention dissemination (spread of interventions) is often addressed later in the research process or is an afterthought in the evaluation process.24 In addition, depictions of scale-up in the public health literature have tended to oversimplify the process25 and have not adequately addressed the complexities involved.26 Early engagement of stakeholders in the research process enhances research-practice translation of interventions,27 can mitigate or reduce the impact of dissemination challenges,27 and potentially increases the likelihood of successful scale-up.28 Given the ongoing lack of evidence for the successful implementation and scale-up of efficacious interventions into practice,22 interventions designed and evaluated with the ‘end application’ in mind may be more likely to achieve sustained population level impact. Nonetheless, the potential for intervention scale-up is rarely considered early in intervention design in this field.

    There are numerous diverse challenges that impact successful scaling of interventions, many of which are outside the control of researchers (such as political climate) or may not be experienced until later in the scale-up process (such as contextual changes in the delivery setting). One aspect within the control of researchers is the design of interventions to ensure their delivery mode enhances widespread reach,29 that they are closely aligned with intended delivery context,30 and that implementation in practice does not exacerbate disparities in health.31 Mobile health (mHealth) is one delivery mode that has the advantage of maximising reach across geographical and socioeconomic groups and has strong potential for scalability and cost-effectiveness.32 33 Use of mHealth strategies (eg, applications (apps) accessed on a mobile phone) are well established in the field of behaviour change but have not been widely trialled in early childhood.34 Systematic reviews of parenting programmes addressing a range of outcomes and delivered online, conclude that web-based delivery for both guided and self-guided interventions result in positive outcomes for parents and children.35 36 A systematic review37 of child and adolescent obesity prevention and treatment interventions using eHealth delivery (ie, mHealth or other electronic access such as via the internet or email) identified eight studies, all including physical activity or screen time targets, alongside diet. All targeted children older than 5 years. Only one study was solely eHealth, with the others also incorporating more traditional delivery modes such as telephone or face-to-face counselling, precluding conclusions on the benefit of eHealth.

    A systematic review38 of web-based interventions to change parent feeding practices for children up to 12 years of age identified seven studies that included children younger than 5 years, one targeting parents of infants, and the remainder focused on preschool children. Some of these studies had a broader focus (eg, obesity prevention) and hence also included one or more movement behaviours in their intervention. That review concluded that there was promise with this delivery mode, but current studies were predominantly of low quality with small sample sizes.38 Given parents of young children often experience time and other logistical barriers to participation in traditional intervention programmes, mHealth strategies offer the potential for good engagement. The few interventions in this population that have utilised mHealth delivery have shown high feasibility and acceptability for parents.39 40

    Aims

    This paper presents the protocol for a trial, the primary aim of which is to assess the efficacy of the Let’s Grow mHealth intervention, a purpose designed mobile web application for parents, to improve the composition of movement behaviours in 2-year-old children at conclusion of the 12-month intervention.

    Secondary aims are to assess the:

    • Maintenance of intervention effects 1-year post intervention.

    • Cost-effectiveness of Let’s Grow measured against current practice (ie, no intervention).

    • Potential mechanisms of behavioural change via hypothesised mediating pathways (eg, parenting practices, parenting confidence).

    • Potential translation and scalability of the intervention into real-world practice.

    Methods and analysis

    Trial design overview

    A randomised controlled trial will test the effects of the 12-month parental support mHealth intervention, Let’s Grow (figure 1). The control group will receive usual care plus electronic newsletters on unrelated topics for continued engagement. Data will be collected at baseline (T0), mid-intervention (T1; 6 months post baseline), at intervention conclusion (T2; 12 months post baseline) and 1-year post intervention (T3; 2 years post baseline). The trial will run from February 2021 to approximately December 2023. Consolidated Standards of Reporting Trials guidelines41 will be followed.

    Figure 1
    Figure 1

    Let’s Grow CONSORT flow chart. CONSORT, Consolidated Standards of Reporting Trials.

    Patient and public involvement

    The intervention was designed and refined with end user (parent) input and piloting prior to commencement of the trial described in this protocol.

    Theoretical framework

    Intervention content was designed using Michie’s Behaviour Change Wheel42 which identifies sources of behaviour (ie, mediators) that the intervention will target across the COM-B domains of: (1) Capability—knowledge, tools, skills; (2) Opportunity—the things that make a behaviour possible and (3) Motivation—goals. Appropriate behaviour change techniques were identified from the CALO-RE taxonomy of behaviour change techniques43 to target each mediator. Identification of target mediators was informed by Social Cognitive Theory-Family Perspective44 and the Family Ecological Model.45 These theories recognise the interplay of parent and child cognitions and behaviours within the family environment, and the multiple influences on child health behaviours: individual (eg, child age, temperament), family (eg, siblings, socioeconomic position), and community domains (eg, access to parks). They are centred around parenting influences (eg, modelling, shaping by rewards and rules, accessibility, knowledge, beliefs).

    An important aspect of this intervention’s design is that it mirrors real world complexity by using nonlinear dynamical approaches (chaos theory, complex dynamic systems).46 47 These approaches operate on the premise that behaviour and behaviour change are nonlinear and unpredictable.48 While health interventions are typically delivered in a predictable linear manner (eg, weekly sessions), the Let’s Grow intervention will be delivered in a dynamic fashion, responsive to individual family situations and parent choice. The aim is to maximise opportunities to target parents during periods when they are most receptive to behaviour change messages (teachable moments).

    Intervention condition

    The Let’s Grow intervention will be delivered via a mobile web app with linked SMS notifications. Intervention content was developed based on the best available evidence, informed by the investigators’ experience with behaviour change interventions in early childhood populations16 34 40 49 50 and with input from end users on iterations of the app. It incorporates practical advice, established behaviour change techniques (eg, goal setting, self-monitoring), and tips and tools to assist parents to improve their child’s physical activity, sleep and sedentary behaviour. This includes information about the behaviours, government guidelines, ideas for promoting each of the behaviours, parenting strategies, and tasks to put participant learning into action, for example, quizzes, goal setting, self-monitoring, creation of routines, strategy implementation, sharing ideas and successes. Information is presented with low literacy requirements using text, images, infographics, animations and live action videos.

    After completing a short introduction module to give them an overview of the intervention and how to use the app, participants will work through eight modules in their own time over 12 months (table 1). They can choose the order and speed in which they undertake the modules. Once they commence a module, they will be unable to open another new module until it has been completed; however, completed modules remain accessible at all times, and key information is freely accessible via the toolkit (described below). This format is designed to allow the behaviour change activities, with linked SMS notifications, to be engaged in without the confusion that could arise by SMS notifications coming from multiple modules concurrently.

    View this table:
    Table 1

    Intervention content

    Within modules, the format is also non-linear with a range of tabs linking to information and activities that participants can undertake in the order of their choosing. Modules also provide an additional ‘burst’ of SMS notifications not linked to specific activities (up to three per module) while that module is active, providing tips and links to push participants back to module content. Once all modules are completed, ‘refresher’ notifications (one per week) push participants back to the app. SMS notifications will be personalised to include parent and child names and, where appropriate, personalised information related to their behaviour change activity (eg, the chosen goal in a goal setting activity). The app also contains a toolkit which stores all videos from completed modules and routines that parents have created, as well as additional information on play, screen time, sleep, parents’ own health behaviours, parenting strategies, tantrums, behaviour change resources, ideas for active play and screen free activities. A social forum using randomly generated usernames (or parents can choose their own but are encouraged not to use real names) gives parents the opportunity to anonymously share ideas and connect with other participants should they wish to. The forum will be monitored by the research team to ensure appropriate use and moderated if necessary. Any moderation will be documented. A frequently asked questions section provides information on use of the app (eg, turning video captions on or off, changing password) and study information (eg, how to pause study involvement, how to wear monitors). A personal profile section allows participants to change their time zone and the preferred time of SMS notifications as well as add a secondary user (another carer) to have view-only access to the app (ie, second users can see information but not undertake the embedded behaviour change activities). Finally, a favourites section allows users to curate their own favourite information from the app by clicking a heart button that appears next to content. Selected content then concurrently appears in the favourites tab, automatically sorted into folders for different information types, for example, videos, activities, toolkit.

    Intervention adherence will be monitored by app usage data. Automated notifications will be sent after predetermined periods of app inactivity (eg, after 2 and 3 weeks of inactivity an SMS will be sent; after 5 weeks of inactivity an email will be sent) to encourage participants to re-engage and complete the programme. If there is continued inactivity after the automated notifications, research staff will contact participants by phone to encourage continuation of the programme. Participants have the option to pause the programme during non-receptive times, for example, if they are going on holidays or dealing with personal issues.

    Control condition

    The control group will continue with any usual care (eg, key ages and stages visits with maternal, child and family health nurses) and will not have access to the Let’s Grow app. They will receive eight electronic bulletins (Toddler Tips) on unrelated topics (eg, basic child first aid, language development, toilet training) delivered approximately every 6 weeks across the 12-month intervention period for cohort engagement. This method has been successful in ensuring high retention in a similar population.51

    Inclusion criteria

    Parents (aged 18+ years) with a child aged 22–35 months will be eligible. This is a key time, commencing when children are ambulant (walking) and encompassing a period of rapid physical and cognitive development. Other inclusion criteria are that the parent resides in Australia, has a mobile phone that can access the internet, can read English, and the child is walking independently.

    Exclusion criteria

    Child has been diagnosed with or is receiving treatment for a sleep disorder.

    Recruitment

    Participants will be recruited nationally, with recruitment open to all eligible parents residing anywhere in Australia. Recruitment strategies will include social media, for example, Facebook, Instagram, parenting blogs and snowball recruitment. These methods have been shown to result in equivalent participant demographics to traditional recruitment methods (eg, face to face) in this population.32 Initial online screening prior to providing consent will ensure participants meet eligibility criteria. Recruitment commenced in February 2021.

    Randomisation

    Randomisation after baseline assessment will be on a 1:1 ratio stratified by geographical location (urban or outer/remote for each of the 8 Australian states/territories; 16 strata). The rationale for stratification is that background health services differ by location and are known to influence effectiveness of interventions.52 The random allocation sequences were computer generated in advance and embedded in REDCap (Vanderbilt, USA), ensuring allocation concealment.

    Measures

    Participants will be assessed pre-randomisation (baseline; T0), mid-way through the intervention (6 months post baseline; T1), at completion of the 12-month intervention phase (T2) and 1-year post intervention (T3). The self-nominated main carer of the child will complete proxy reports on their child’s behaviour as well as provide information on themselves and their family demographics. The partner of the main carer or other parent of the child will also be invited to participate (via the main carer) and provide information on their own behaviour. The measures included at each time point are outlined in online supplemental table.

    Movement behaviour data (physical activity, sedentary behaviour and sleep)

    The primary outcome is daily proportion of time spent in physical activity, sedentary behaviour and sleep at completion of the intervention (T2). At T0, T2 and T3 child movement behaviours will be concurrently assessed using ActiGraph GT3X+ (ActiGraph, Pensacola, USA) accelerometers, worn at the hip continuously for 24 hours across 8 days. Rewear will be requested where monitors are returned with less than 7.4 waking hours per day across a minimum of 4 days recorded. Accelerometers are the gold standard for assessing free living movement and have been validated with children as young as 16 months.53 Hip worn accelerometers provide a valid measure of sleep duration in children.54 Data will be recorded in 5 s epochs to capture the sporadic nature of young children’s movement and 20 min of consecutive zero counts will be considered non-wear time and removed from analyses.55 Sedentary behaviour53 and different intensities of physical activity56 will be estimated with age-specific movement count thresholds. Sleep duration will be estimated using a sleep-wake detection algorithm developed in MATLAB (MathWorks, Natick, Massachusetts, USA) that automatically finds and scores daytime naps and overnight sleep after age-appropriate approximations for evening sleep and morning wake times are entered into the MATLAB script. Some manual processing may be required for those with unpredictable times. The algorithm has been shown to have good agreement with parental sleep diaries.55 Monitors will be initialised and posted to families with stickers identifying who should wear each monitor. Text messages or email will remind parents to ensure their child wears the monitor and to post it back at the end of the monitoring period. The monitors will be accompanied by a booklet providing instructions on how to fit the device, tips for wear and how to return the device. The booklet will also collect daily parent-reported data during monitor wear for non-parent care and single questions on child self-regulation and ease of parenting that day (covariates).

    Additional contextual information on child movement behaviours will be collected by main carer parent report via REDCap (online supplemental table). This information will include parent report of the amount of time their child spends in a range of physical activities (eg, indoor and outdoor active play, high energy play), screen and non-screen based sedentary behaviours (eg, stroller/pram, watching shows, playing electronic games) and sleeping (eg, usual bedtime and wake time, night waking, sleep routines). Survey items are based on established measures where available. Purpose designed items will undergo reliability testing in a separate sample.

    Potential mediators

    Targeted mediators of the intervention including movement behaviour-related parent knowledge, efficacy/confidence, family rules and routines, co-participation and home environment, as well as child motor skills, general parenting and child behaviour will be assessed at all time points by parent report using existing reliable instruments (online supplemental table). Parent modelling of behaviours will be device assessed at T0, T2 and T3 by parents wearing GT3X accelerometers concurrently with their child. In addition, parents will report on their own movement behaviours in relation to adult guidelines. This information will help identify possible pathways through which the intervention had an effect.

    Demographic information and covariates

    Standard demographic and socioeconomic information will be collected via parent report. In addition, data will be collected on potential covariates including child health conditions, temperament, dietary intake (food frequency questionnaire), birth weight and length, parent coping, concern with child movement behaviours, parent height and weight, and maternal pregnancy status (online supplemental table). Child height and weight data will also be collected through the online survey. Parents will be asked to upload a picture of their child’s health record or copy the information into the survey if they have had measures recorded by their maternal child health nurse or another health professional recently. If they do not have recent measures, they will be asked to measure their child’s height (against a wall) and weight (on bathroom scales) at home with instructions provided.

    Secondary outcomes

    Measures to be used for the economic evaluation include parent report of health service utilisation and cost,57 and time spent seeking information online. In addition, a log of researcher time administering the intervention (eg, monitoring the social forum) and costs associated with this will be recorded by the research team (online supplemental table).

    For the process evaluation, we will ask all participants to report how they heard about the study in the baseline parent survey. In addition, at T1, T2 and T3 participants in the intervention group will be asked to provide feedback on their engagement, relevance and satisfaction with the intervention. Analytics will be collected directly from the app to provide information on individual participants’ usage across the study (online supplemental table).

    Sample size

    The aim is to recruit 1100 families. Assuming 15% attrition rate at T2, 25% attrition at T352 and 75% of children providing valid accelerometer data at each time point, approximately 700 children at T2 and 620 at T3 will provide valid accelerometry data. Power calculations are based on Hotelling’s T-squared test statistic, α=0.05, software PASS V.14.0.9 (NCSS). Accelerometry data for children 3.5 years and 5 years16 were used to estimate time-use in the control group and the 2×2 variance-covariance matrix.

    Primary outcome (daily proportion of time spent in physical activity, sedentary behaviour and sleep at T2)

    The target sample size of 700 (350 per group) will achieve 89% power to detect a 0.031 increase in physical activity and 0.038 decrease in sedentary time measured in the log-ratio scale (covariance matrix, Embedded Image). Thus, the study is powered to detect small changes translating to +9 min in physical activity, –13 min in sedentary time, and +4 min in sleep when the proportion of time spent in these behaviours in the control group is 16.9%, 28.1% and 55.0% respectively (as for 3.5 years in a prior study).16

    Secondary outcome (daily proportion of time spent in physical activity, sedentary behaviour and sleep at T3)

    The target sample size of 620 children (310 per group) at T3 would achieve 93% power to detect the same changes in movement behaviour minutes as for the primary outcome (Embedded Image) when the proportion of time spent in physical activity, sedentary behaviour and sleep in the control group is 16.4%, 29.2% and 54.4% respectively (as for 5 years in a previous study).16 Of note, power is larger at T3 for the same effects measured in minutes even with a smaller sample size due to the change in the distribution of time-use and the variability in the 5 years data compared with that of 3.5 years.

    Statistical analysis

    All analyses will be conducted on an intention-to-treat basis with the analyst blinded to allocation. Time-use data (primary outcome) are constrained to a total of 24 hours precluding the application of standard statistical techniques on the raw data. Compositional analysis will be undertaken to address the primary aim.58 The fraction of daily time spent in physical activity (x1), sedentary behaviour (x2) and sleep (x3) will be transformed to new variables y1=log(x1/x3) and y2=log(x1/x3) using the proportion of time spent sleeping as the reference. This transformation translates the vector of proportions (x1, x2, x3), which sum to 1, into a bidimensional vector (y1, y2) whose components are no longer constrained and can be analysed using standard multivariate methods.59 The intervention effect on the log-ratio transformed data will be estimated using a multivariate linear mixed-effect model with trial arm as fixed effect and a 2×2 unstructured variance-covariance matrix to account for the correlation between components’ proportions.60 To facilitate interpretation, estimated mean log-ratios will be back transformed into proportion of time in each behaviour (minutes/day). To estimate the longitudinal effect of the intervention on the log-ratio transformed proportions, a multivariate linear mixed-effect model will be fitted with trial arm, time (T0, T2, T3) and interaction arm×time as fixed effects, child as a random effect to account for the repeated measures, and a 2×2 unstructured variance-covariance matrix. Classical and robust imputation algorithms for dealing with missing values in compositional data will be used.61 62

    Intervention effects on parent reported time in various movement behaviours will be estimated using generalised estimating equation models with link and distribution selected according to the outcome. All models will include trial arm, time and arm×time effects. A causal mediation analysis, using a counterfactual framework, will explore whether the effects of the intervention on each individual time fraction (active, sedentary, sleep) is mediated by the targeted parenting practices and attitudes. The mediated (indirect) effect will be computed through G-estimation incorporating confounders of the mediator-outcome association.63

    Economic evaluation

    Incremental cost-effectiveness analysis will be undertaken to determine whether the intervention represents ‘value for money’ compared with current practice (ie, usual care with no access to the Let’s Grow app). This will address technical efficiency (ie, ‘how to do it’) by analysing the net cost and net health benefit of the intervention, and will allow for determination of key intervention design features and their associated cost drivers through a trial-based economic evaluation. The cost-effectiveness analysis will be conducted from both a funder and a limited societal perspective, using detailed pathway analysis to specify all relevant intervention activities and costs. Resource use will be measured using unit costs drawn from trial data and published sources for the 2021 reference year. Cost data and trial outcomes data will be combined in a cost-consequence analysis, reporting a range of incremental cost-effectiveness ratios including cost per minute of screen-time saved, cost per additional unit of sleep and cost per metabolic equivalent task minute gained.

    A modelled economic evaluation will also be undertaken for all movement behaviour outcomes together and separately by extending the target population, time horizon and decision context of the intervention. An existing multi-state Markov model64 will be used to evaluate the intervention’s cost-effectiveness (in terms of cost per health-adjusted life-year (HALY) saved), assuming it was delivered to the eligible Australian population and the observed intervention effect was extrapolated over the cohort’s lifetime. In addition to incremental costs of the intervention (measured against the comparator), incremental cost offsets attributable to disease prevention over the life course will be reported. The commonly accepted reference threshold for cost-effectiveness of $A50 000 per HALY saved will be used.

    Standard discounting will be applied to both costs and outcomes. Simulation modelling using the @RISK and Ersatz software packages will be used to calculate 95% uncertainty intervals (median, 2.5 and 97.5 percentiles) around the epidemiological probabilities and cost estimates. Sensitivity analyses will be undertaken, varying key input parameters into the economic evaluation and gauging overall impact on cost-effectiveness results.

    Process evaluation

    Web app analytics will provide information on frequency and duration of app access, pages visited, order of module completion and activities and modules completed to provide a measure of parent engagement and dose of intervention received by each participant. We will use an engagement index adapted from the Web Analytics Demystified visitor engagement index65 that measures five subindices: click depth; loyalty; interaction; recency; and feedback. This information will be supplemented by quantitative data collected from all intervention group participants and qualitative data from interviews with a subsample of approximately 20 intervention participants on acceptability, satisfaction and relevance of the programme. Analysis of process evaluation data will occur prior to trial outcome data to minimise bias.66

    Scalability evaluation

    A novel addition to this study is inclusion of a prospective scalability evaluation. Our evaluation will be guided by the RE-AIM framework,67 a global framework to evaluate the translation outcomes of health promotion programmes and the PRACTIS (PRACTical planning for Implementation and Scale-up) guide,68 a framework for planning future implementation and scale-up with key stakeholders across multiple levels of the intended delivery system. Prospective scalability will be assessed based on: (1) parent acceptability and engagement with the intervention, (2) associations between intervention fidelity (participant adherence/level of engagement with the intervention as assessed by the engagement index) and impact on behavioural outcomes (effectiveness) and (3) parent and stakeholder (eg, government, family-facing organisations) perceptions of factors related to the future implementation and scale-up (such as individual and system level factors that may enhance or impede wider translation). These will be captured during interviews with approximately 20 participants and the Let’s Grow Stakeholder Group and an online PRACTIS workshop with the Stakeholder Group.68 Data will be transcribed verbatim and thematically analysed using NVivo V.12 software (QSR International, Melbourne, Australia).

    Discussion

    This protocol addresses primary prevention of suboptimal movement behaviours, which impact young children’s current and future health and well-being, as called for in the recent WHO Report on Ending Childhood Obesity.1 The Let’s Grow intervention incorporates a number of innovative features. It adopts a time-use epidemiology approach, which has recently been embodied in WHO and numerous national movement behaviour guidelines conceptualising the 24-hour day as consisting of time spent in physical activity, sedentary behaviour and sleep,10 and substitution as key to behaviour change strategies. It recognises that behaviour change does not occur in a linear manner, and thus is designed to be dynamic and non-linear, reacting to differing family situations and teachable moments. Furthermore, this tailored non-linear approach aligns with the dynamic way in which children develop motor skills and physical literacy over time through exposure to diverse physical activity opportunities,47 thereby assisting children on their journey to lead an active life. Given the novel work with stakeholders to address implementation potential throughout the development and delivery of Let’s Grow, if effective, it has the potential to be offered widely to parents of young children, to support them in promoting healthy movement behaviours from early in life.

    Ethics and dissemination

    The trial has received ethical approval from the Deakin University Human Ethics Committee (2020-077). After eligibility screening, the main carer will provide consent via an online form for their own and their child’s participation in the study. Partners will provide consent for their own participation.

    Trial findings will be disseminated via peer-reviewed publications, presentations at scientific and professional conferences, and via social and traditional media. In addition, findings will be disseminated directly to stakeholders involved in the scale-up evaluation.

    Ethics statements

    Patient consent for publication

    References

    1. ↵
      1. World Health Organization
      . Report of the Commission on ending childhood obesity, 2016. Available: https://www.who.int/publications/i/item/9789241510066
    2. ↵
      1. Carson V,
      2. Lee E-Y,
      3. Hewitt L, et al
      . Systematic review of the relationships between physical activity and health indicators in the early years (0-4 years). BMC Public Health 2017;17:854.doi:10.1186/s12889-017-4860-0pmid:http://www.ncbi.nlm.nih.gov/pubmed/29219090
      OpenUrlPubMed
    3. ↵
      1. Chaput J-P,
      2. Gray CE,
      3. Poitras VJ, et al
      . Systematic review of the relationships between sleep duration and health indicators in the early years (0-4 years). BMC Public Health 2017;17:855.doi:10.1186/s12889-017-4850-2pmid:http://www.ncbi.nlm.nih.gov/pubmed/29219078
      OpenUrlPubMed
    4. ↵
      1. Poitras VJ,
      2. Gray CE,
      3. Janssen X, et al
      . Systematic review of the relationships between sedentary behaviour and health indicators in the early years (0-4 years). BMC Public Health 2017;17:868.doi:10.1186/s12889-017-4849-8pmid:http://www.ncbi.nlm.nih.gov/pubmed/29219092
      OpenUrlPubMed
    5. ↵
      1. Carter PJ,
      2. Taylor BJ,
      3. Williams SM, et al
      . Longitudinal analysis of sleep in relation to BMI and body fat in children: the flame study. BMJ 2011;342:d2712.doi:10.1136/bmj.d2712pmid:http://www.ncbi.nlm.nih.gov/pubmed/21622518
      OpenUrlAbstract/FREE Full Text
    6. ↵
      1. Santos R,
      2. Zhang Z,
      3. Pereira JR, et al
      . Compliance with the Australian 24-hour movement guidelines for the early years: associations with weight status. BMC Public Health 2017;17:867.doi:10.1186/s12889-017-4857-8pmid:http://www.ncbi.nlm.nih.gov/pubmed/29219095
      OpenUrlPubMed
    7. ↵
      1. Cliff DP,
      2. McNeill J,
      3. Vella SA, et al
      . Adherence to 24-hour movement guidelines for the early years and associations with social-cognitive development among Australian preschool children. BMC Public Health 2017;17:857.doi:10.1186/s12889-017-4858-7pmid:http://www.ncbi.nlm.nih.gov/pubmed/29219104
      OpenUrlPubMed
    8. ↵
      1. Okely AD,
      2. Ghersi D,
      3. Hesketh KD, et al
      . A collaborative approach to adopting/adapting guidelines - The Australian 24-Hour Movement Guidelines for the early years (Birth to 5 years): an integration of physical activity, sedentary behavior, and sleep. BMC Public Health 2017;17:869.doi:10.1186/s12889-017-4867-6pmid:http://www.ncbi.nlm.nih.gov/pubmed/29219094
      OpenUrlPubMed
    9. ↵
      1. Tremblay MS
      . Introducing 24-hour movement guidelines for the early years: a new paradigm gaining momentum. J Phys Act Health 2020;17:92–5.doi:10.1123/jpah.2019-0401pmid:http://www.ncbi.nlm.nih.gov/pubmed/31711035
      OpenUrlPubMed
    10. ↵
      1. Rollo S,
      2. Antsygina O,
      3. Tremblay MS
      . The whole day matters: understanding 24-hour movement guideline adherence and relationships with health indicators across the lifespan. J Sport Health Sci 2020;9:493–510.doi:10.1016/j.jshs.2020.07.004pmid:http://www.ncbi.nlm.nih.gov/pubmed/32711156
      OpenUrlPubMed
    11. ↵
      1. Meredith-Jones K,
      2. Galland B,
      3. Haszard J, et al
      . Do young children consistently meet 24-h sleep and activity guidelines? A longitudinal analysis using actigraphy. Int J Obes 2019;43:2555–64.doi:10.1038/s41366-019-0432-ypmid:http://www.ncbi.nlm.nih.gov/pubmed/31477783
      OpenUrlPubMed
    12. ↵
      1. Delisle Nyström C,
      2. Alexandrou C,
      3. Henström M, et al
      . International study of movement behaviors in the early years (SUNRISE): results from SUNRISE Sweden's pilot and COVID-19 study. Int J Environ Res Public Health 2020;17. doi:doi:10.3390/ijerph17228491. [Epub ahead of print: 16 11 2020].pmid:http://www.ncbi.nlm.nih.gov/pubmed/33207786
    13. ↵
      1. Brown T,
      2. Moore TH,
      3. Hooper L, et al
      . Interventions for preventing obesity in children. Cochrane Database Syst Rev 2019;7:CD001871.doi:10.1002/14651858.CD001871.pub4pmid:http://www.ncbi.nlm.nih.gov/pubmed/31332776
      OpenUrlPubMed
    14. ↵
      1. De Bourdeaudhuij I,
      2. Verbestel V,
      3. De Henauw S, et al
      . Behavioural effects of a community-oriented setting-based intervention for prevention of childhood obesity in eight European countries. main results from the IDEFICS study. Obes Rev 2015;16 Suppl 2:30–40.doi:10.1111/obr.12347pmid:http://www.ncbi.nlm.nih.gov/pubmed/26707014
      OpenUrlPubMed
    15. ↵
      1. Michels N,
      2. De Henauw S,
      3. Eiben G, et al
      . Effect of the IDEFICS multilevel obesity prevention on children's sleep duration. Obes Rev 2015;16 Suppl 2:68–77.doi:10.1111/obr.12327pmid:http://www.ncbi.nlm.nih.gov/pubmed/26707017
      OpenUrlPubMed
    16. ↵
      1. Taylor BJ,
      2. Gray AR,
      3. Galland BC, et al
      . Targeting sleep, food, and activity in infants for obesity prevention: an RCT. Pediatrics 2017;139:e20162037.doi:10.1542/peds.2016-2037pmid:http://www.ncbi.nlm.nih.gov/pubmed/28242860
      OpenUrlAbstract/FREE Full Text
    17. ↵
      1. Henriksson H,
      2. Alexandrou C,
      3. Henriksson P, et al
      . MINISTOP 2.0: a smartphone APP integrated in primary child health care to promote healthy diet and physical activity behaviours and prevent obesity in preschool-aged children: protocol for a hybrid design effectiveness-implementation study. BMC Public Health 2020;20:1756.doi:10.1186/s12889-020-09808-wpmid:http://www.ncbi.nlm.nih.gov/pubmed/33228572
      OpenUrlPubMed
    18. ↵
      1. Milat AJ,
      2. King L,
      3. Bauman AE, et al
      . The concept of scalability: increasing the scale and potential adoption of health promotion interventions into policy and practice. Health Promot Int 2013;28:285–98.doi:10.1093/heapro/dar097pmid:http://www.ncbi.nlm.nih.gov/pubmed/22241853
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. McCrabb S,
      2. Lane C,
      3. Hall A, et al
      . Scaling-up evidence-based obesity interventions: a systematic review assessing intervention adaptations and effectiveness and quantifying the scale-up penalty. Obes Rev 2019;20:964–82.doi:10.1111/obr.12845pmid:http://www.ncbi.nlm.nih.gov/pubmed/30868745
      OpenUrlPubMed
    20. ↵
      1. Harden SM,
      2. Johnson SB,
      3. Almeida FA, et al
      . Improving physical activity program adoption using integrated research-practice partnerships: an effectiveness-implementation trial. Transl Behav Med 2017;7:28–38.doi:10.1007/s13142-015-0380-6pmid:http://www.ncbi.nlm.nih.gov/pubmed/28299746
      OpenUrlPubMed
    21. ↵
      1. Scheirer MA,
      2. Dearing JW
      . An agenda for research on the sustainability of public health programs. Am J Public Health 2011;101:2059–67.doi:10.2105/AJPH.2011.300193pmid:http://www.ncbi.nlm.nih.gov/pubmed/21940916
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. Reis RS,
      2. Salvo D,
      3. Ogilvie D, et al
      . Scaling up physical activity interventions worldwide: stepping up to larger and smarter approaches to get people moving. Lancet 2016;388:1337–48.doi:10.1016/S0140-6736(16)30728-0pmid:http://www.ncbi.nlm.nih.gov/pubmed/27475273
      OpenUrlCrossRefPubMed
    23. ↵
      1. World Health Organization
      . Global action plan on physical activity 2018-2030: more active people for a healthier world, 2018. Available: https://apps.who.int/iris/handle/10665/272722
    24. ↵
      1. Bopp M,
      2. Saunders RP,
      3. Lattimore D
      . The tug-of-war: fidelity versus adaptation throughout the health promotion program life cycle. J Prim Prev 2013;34:193–207.doi:10.1007/s10935-013-0299-ypmid:http://www.ncbi.nlm.nih.gov/pubmed/23526141
      OpenUrlCrossRefPubMed
    25. ↵
      1. Koorts H,
      2. Rutter H
      . A systems approach to scale-up for population health improvement. Health Res Policy Syst 2021;19:27.doi:10.1186/s12961-021-00679-0pmid:http://www.ncbi.nlm.nih.gov/pubmed/33648525
      OpenUrlPubMed
    26. ↵
      1. Koorts H,
      2. Cassar S,
      3. Salmon J, et al
      . Mechanisms of scaling up: combining a realist perspective and systems analysis to understand successfully scaled interventions. Int J Behav Nutr Phys Act 2021;18:42.doi:10.1186/s12966-021-01103-0pmid:http://www.ncbi.nlm.nih.gov/pubmed/33752681
      OpenUrlPubMed
    27. ↵
      1. Klesges LM,
      2. Estabrooks PA,
      3. Dzewaltowski DA, et al
      . Beginning with the application in mind: designing and planning health behavior change interventions to enhance dissemination. Ann Behav Med 2005;29 Suppl:66–75.doi:10.1207/s15324796abm2902s_10pmid:http://www.ncbi.nlm.nih.gov/pubmed/15921491
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. World Health Organization
      . Beginning with the end in mind: planning pilot projects and otherprogrammatic research for successful scaling up, 2011. Available: https://www.who.int/reproductivehealth/publications/strategic_approach/9789241502320/en/
    29. ↵
      1. World Health Organization
      . The maps toolkit: mHealth assessment and planning for scale, 2015. Available: https://apps.who.int/iris/handle/10665/185238
    30. ↵
      1. Zamboni K,
      2. Schellenberg J,
      3. Hanson C, et al
      . Assessing scalability of an intervention: why, how and who? Health Policy Plan 2019;34:544–52.doi:10.1093/heapol/czz068pmid:http://www.ncbi.nlm.nih.gov/pubmed/31365066
      OpenUrlCrossRefPubMed
    31. ↵
      1. Baumann AA,
      2. Cabassa LJ
      . Reframing implementation science to address inequities in healthcare delivery. BMC Health Serv Res 2020;20:190.doi:10.1186/s12913-020-4975-3pmid:http://www.ncbi.nlm.nih.gov/pubmed/32164706
      OpenUrlPubMed
    32. ↵
      1. Laws RA,
      2. Litterbach E-KV,
      3. Denney-Wilson EA, et al
      . A comparison of recruitment methods for an mHealth intervention targeting mothers: lessons from the growing healthy program. J Med Internet Res 2016;18:e248–e48.doi:10.2196/jmir.5691pmid:http://www.ncbi.nlm.nih.gov/pubmed/27634633
      OpenUrlPubMed
    33. ↵
      1. Brown V,
      2. Tran H,
      3. Downing KL, et al
      . A systematic review of economic evaluations of web-based or telephone-delivered interventions for preventing overweight and obesity and/or improving obesity-related behaviors. Obes Rev 2021;22:e13227.doi:10.1111/obr.13227pmid:http://www.ncbi.nlm.nih.gov/pubmed/33763956
      OpenUrlPubMed
    34. ↵
      1. Downing KL,
      2. Salmon J,
      3. Hinkley T, et al
      . Feasibility and Efficacy of a Parent-Focused, Text Message-Delivered Intervention to Reduce Sedentary Behavior in 2- to 4-Year-Old Children (Mini Movers): Pilot Randomized Controlled Trial. JMIR Mhealth Uhealth 2018;6:e39.doi:10.2196/mhealth.8573pmid:http://www.ncbi.nlm.nih.gov/pubmed/29426816
      OpenUrlPubMed
    35. ↵
      1. Nieuwboer CC,
      2. Fukkink RG,
      3. Hermanns JMA
      . Peer and professional parenting support on the Internet: a systematic review. Cyberpsychol Behav Soc Netw 2013;16:518–28.doi:10.1089/cyber.2012.0547pmid:http://www.ncbi.nlm.nih.gov/pubmed/23659725
      OpenUrlCrossRefPubMed
    36. ↵
      1. Nieuwboer CC,
      2. Fukkink RG,
      3. Hermanns JMA
      . Online programs as tools to improve parenting: a meta-analytic review. Child Youth Serv Rev 2013;35:1823–9.doi:10.1016/j.childyouth.2013.08.008
      OpenUrlCrossRef
    37. ↵
      1. Hammersley ML,
      2. Jones RA,
      3. Okely AD
      . Parent-Focused childhood and adolescent overweight and obesity eHealth interventions: a systematic review and meta-analysis. J Med Internet Res 2016;18:e203.doi:10.2196/jmir.5893pmid:http://www.ncbi.nlm.nih.gov/pubmed/27443862
      OpenUrlPubMed
    38. ↵
      1. Gomes AI,
      2. Pereira AI,
      3. Roberto MS, et al
      . Changing parental feeding practices through web-based interventions: a systematic review and meta-analysis. PLoS One 2021;16:e0250231.doi:10.1371/journal.pone.0250231pmid:http://www.ncbi.nlm.nih.gov/pubmed/33909666
      OpenUrlPubMed
    39. ↵
      1. Nezami BT,
      2. Ward DS,
      3. Lytle LA, et al
      . A mHealth randomized controlled trial to reduce sugar-sweetened beverage intake in preschool-aged children. Pediatr Obes 2018;13:668–76.doi:10.1111/ijpo.12258pmid:http://www.ncbi.nlm.nih.gov/pubmed/29119719
      OpenUrlPubMed
    40. ↵
      1. Nyström CD,
      2. Sandin S,
      3. Henriksson P, et al
      . Mobile-based intervention intended to stop obesity in preschool-aged children: the MINISTOP randomized controlled trial. Am J Clin Nutr 2017;105:ajcn150995–35.doi:10.3945/ajcn.116.150995pmid:http://www.ncbi.nlm.nih.gov/pubmed/28446496
      OpenUrlPubMed
    41. ↵
      1. Schulz KF,
      2. Altman DG,
      3. Moher D, et al
      . Consort 2010 statement: updated guidelines for reporting parallel group randomised trials. BMJ 2010;340:c332.doi:10.1136/bmj.c332pmid:http://www.ncbi.nlm.nih.gov/pubmed/20332509
      OpenUrlFREE Full Text
    42. ↵
      1. Michie S,
      2. van Stralen MM,
      3. West R
      . The behaviour change wheel: a new method for characterising and designing behaviour change interventions. Implement Sci 2011;6:42.doi:10.1186/1748-5908-6-42pmid:http://www.ncbi.nlm.nih.gov/pubmed/21513547
      OpenUrlCrossRefPubMed
    43. ↵
      1. Michie S,
      2. Ashford S,
      3. Sniehotta FF, et al
      . A refined taxonomy of behaviour change techniques to help people change their physical activity and healthy eating behaviours: the CALO-RE taxonomy. Psychol Health 2011;26:1479–98.doi:10.1080/08870446.2010.540664pmid:http://www.ncbi.nlm.nih.gov/pubmed/21678185
      OpenUrlCrossRefPubMedWeb of Science
    44. ↵
      1. Taylor WC,
      2. Baranowski T,
      3. Sallis JF
      . Family determinants of childhood physical activity: a social cognitive model. In: Advances in exercise adherence human kinetics. Champaign, IL, 1994.
    45. ↵
      1. Campbell KJ,
      2. Hesketh KD,
      3. Davison KK
      . The role of parents in preventing child overweight and obesity: an ecological approach. In: Obesity epidemiology; from aetiology to public health. 2nd edn. London: Oxford University Press, 2010.
    46. ↵
      1. Resnicow K,
      2. Page SE
      . Embracing chaos and complexity: a quantum change for public health. Am J Public Health 2008;98:1382–9.doi:10.2105/AJPH.2007.129460pmid:http://www.ncbi.nlm.nih.gov/pubmed/18556599
      OpenUrlCrossRefPubMedWeb of Science
    47. ↵
      1. Rudd JR,
      2. Pesce C,
      3. Strafford BW, et al
      . Physical Literacy - a journey of individual enrichment: an ecological dynamics rationale for enhancing performance and physical activity in all. Front Psychol 2020;11:11.doi:10.3389/fpsyg.2020.01904pmid:http://www.ncbi.nlm.nih.gov/pubmed/32849114
      OpenUrlCrossRefPubMed
    48. ↵
      1. Riley WT,
      2. Rivera DE,
      3. Atienza AA, et al
      . Health behavior models in the age of mobile interventions: are our theories up to the task? Transl Behav Med 2011;1:53–71.doi:10.1007/s13142-011-0021-7pmid:http://www.ncbi.nlm.nih.gov/pubmed/21796270
      OpenUrlCrossRefPubMed
    49. ↵
      1. Hesketh KD,
      2. Kuswara K,
      3. Abbott G, et al
      . How to change young children’s physical activity and sedentary behavior: mechanisms of behavior change in the infant cluster randomized controlled trial. Children 2021;8:470.doi:10.3390/children8060470
    50. ↵
      1. Downing KL,
      2. Best K,
      3. Campbell KJ, et al
      . Informing active play and screen time behaviour change interventions for low socioeconomic position mothers of young children: what do mothers want? Biomed Res Int 2016;2016:1–13.doi:10.1155/2016/2139782pmid:http://www.ncbi.nlm.nih.gov/pubmed/28053979
      OpenUrlPubMed
    51. ↵
      1. Hesketh KD,
      2. Salmon J,
      3. McNaughton SA, et al
      . Long-term outcomes (2 and 3.5 years post-intervention) of the INFANT early childhood intervention to improve health behaviors and reduce obesity: cluster randomised controlled trial follow-up. Int J Behav Nutr Phys Act 2020;17:95.doi:10.1186/s12966-020-00994-9pmid:http://www.ncbi.nlm.nih.gov/pubmed/32711523
      OpenUrlPubMed
    52. ↵
      1. Askie LM,
      2. Baur LA,
      3. Campbell K, et al
      . The Early Prevention of Obesity in CHildren (EPOCH) Collaboration--an individual patient data prospective meta-analysis. BMC Public Health 2010;10:728.doi:10.1186/1471-2458-10-728pmid:http://www.ncbi.nlm.nih.gov/pubmed/21106110
      OpenUrlPubMed
    53. ↵
      1. Trost SG,
      2. Fees BS,
      3. Haar SJ, et al
      . Identification and validity of accelerometer cut-points for toddlers. Obesity 2012;20:2317–9.doi:10.1038/oby.2011.364pmid:http://www.ncbi.nlm.nih.gov/pubmed/22173573
      OpenUrlPubMed
    54. ↵
      1. Kinder JR,
      2. Lee KA,
      3. Thompson H, et al
      . Validation of a hip-worn accelerometer in measuring sleep time in children. J Pediatr Nurs 2012;27:127–33.doi:10.1016/j.pedn.2010.11.004pmid:http://www.ncbi.nlm.nih.gov/pubmed/22341191
      OpenUrlCrossRefPubMed
    55. ↵
      1. Galland B,
      2. Meredith-Jones K,
      3. Gray A, et al
      . Criteria for nap identification in infants and young children using 24-h actigraphy and agreement with parental diary. Sleep Med 2016;19:85–92.doi:10.1016/j.sleep.2015.10.013pmid:http://www.ncbi.nlm.nih.gov/pubmed/27198952
      OpenUrlPubMed
    56. ↵
      1. Pate RR,
      2. Almeida MJ,
      3. McIver KL, et al
      . Validation and calibration of an accelerometer in preschool children. Obesity 2006;14:2000–6.doi:10.1038/oby.2006.234pmid:http://www.ncbi.nlm.nih.gov/pubmed/17135617
      OpenUrlPubMed
    57. ↵
      1. Campbell KJ,
      2. Lioret S,
      3. McNaughton SA, et al
      . A parent-focused intervention to reduce infant obesity risk behaviors: a randomized trial. Pediatrics 2013;131:652–60.doi:10.1542/peds.2012-2576pmid:http://www.ncbi.nlm.nih.gov/pubmed/23460688
      OpenUrlAbstract/FREE Full Text
    58. ↵
      1. Aitchison J
      . The statistical analysis of compositional data. London: Chapman Hall, 1986.
    59. ↵
      1. Dumuid D,
      2. Stanford TE,
      3. Martin-Fernández J-A, et al
      . Compositional data analysis for physical activity, sedentary time and sleep research. Stat Methods Med Res 2018;27:3726–38.doi:10.1177/0962280217710835pmid:http://www.ncbi.nlm.nih.gov/pubmed/28555522
      OpenUrlPubMed
    60. ↵
      1. Faes C,
      2. Molenberghs G,
      3. Hens N, et al
      . Analysing the composition of outpatient antibiotic use: a tutorial on compositional data analysis. J Antimicrob Chemother 2011;66 Suppl 6:vi89–94.doi:10.1093/jac/dkr461pmid:http://www.ncbi.nlm.nih.gov/pubmed/22096070
      OpenUrlPubMed
    61. ↵
      1. Hron K,
      2. Templ M,
      3. Filzmoser P
      . Imputation of missing values for compositional data using classical and robust methods. Comput Stat Data Anal 2010;54:3095–107.doi:10.1016/j.csda.2009.11.023
      OpenUrlCrossRefWeb of Science
    62. ↵
      1. Real C,
      2. Ángel Fernández J,
      3. Aboal JR, et al
      . Substituting missing data in compositional analysis. Environ Pollut 2011;159:2797–800.doi:10.1016/j.envpol.2011.05.006pmid:http://www.ncbi.nlm.nih.gov/pubmed/21645952
      OpenUrlPubMed
    63. ↵
      1. Valeri L,
      2. Vanderweele TJ
      . Mediation analysis allowing for exposure-mediator interactions and causal interpretation: theoretical assumptions and implementation with SAS and SPSS macros. Psychol Methods 2013;18:137–50.doi:10.1037/a0031034pmid:http://www.ncbi.nlm.nih.gov/pubmed/23379553
      OpenUrlCrossRefPubMedWeb of Science
    64. ↵
      1. Ananthapavan J,
      2. Sacks G,
      3. Brown V, et al
      . Priority-setting for obesity prevention-The assessing cost-effectiveness of obesity prevention policies in Australia (ACE-Obesity policy) study. PLoS One 2020;15:e0234804.doi:10.1371/journal.pone.0234804pmid:http://www.ncbi.nlm.nih.gov/pubmed/32559212
      OpenUrlPubMed
    65. ↵
      1. Taki S,
      2. Lymer S,
      3. Russell CG, et al
      . Assessing user engagement of an mHealth intervention: development and implementation of the growing healthy APP engagement index. JMIR Mhealth Uhealth 2017;5:e89.doi:10.2196/mhealth.7236pmid:http://www.ncbi.nlm.nih.gov/pubmed/28663164
      OpenUrlPubMed
    66. ↵
      1. Moore GF,
      2. Audrey S,
      3. Barker M, et al
      . Process evaluation of complex interventions: medical Research Council guidance. BMJ 2015;350:h1258.doi:10.1136/bmj.h1258pmid:http://www.ncbi.nlm.nih.gov/pubmed/25791983
      OpenUrlFREE Full Text
    67. ↵
      1. Glasgow RE,
      2. Vogt TM,
      3. Boles SM
      . Evaluating the public health impact of health promotion interventions: the RE-AIM framework. Am J Public Health 1999;89:1322–7.doi:10.2105/AJPH.89.9.1322pmid:http://www.ncbi.nlm.nih.gov/pubmed/10474547
      OpenUrlCrossRefPubMedWeb of Science
    68. ↵
      1. Koorts H,
      2. Eakin E,
      3. Estabrooks P, et al
      . Implementation and scale up of population physical activity interventions for clinical and community settings: the PRACTIS guide. Int J Behav Nutr Phys Act 2018;15:51.doi:10.1186/s12966-018-0678-0pmid:http://www.ncbi.nlm.nih.gov/pubmed/29884236
      OpenUrlPubMed
      1. Michie S,
      2. Richardson M,
      3. Johnston M, et al
      . The behavior change technique taxonomy (V1) of 93 hierarchically clustered techniques: building an international consensus for the reporting of behavior change interventions. Ann Behav Med 2013;46:81–95.doi:10.1007/s12160-013-9486-6pmid:http://www.ncbi.nlm.nih.gov/pubmed/23512568
      OpenUrlCrossRefPubMed

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • Twitter @KylieHesketh, @DrKDowning, @Harriet_Koorts, @Vicki_BBB, @JessHainesPhD, @ProfKCampbell, @LisaBarnettPhD, @ValLCarson, @profjsalmon

    • Contributors All authors are named investigators on the funding application for this trial and contributed to the writing of this protocol. KDH conceived the idea for the study and led development of the intervention and efficacy evaluation and writing of this protocol. KLD, BCG, JN and JS contributed to development of the intervention content. MA led development of the web application. LO developed the sample size and analysis plan. HK developed the scalability evaluation. VB and MM developed the economic evaluation. KLD, BCG, JN, RT, LO, JH, KC, LB, ML, VC and JS contributed to design of the efficacy evaluation.

    • Funding This trial was supported by the National Health and Medical Research Council (GNT 1162980). In addition, VB is supported by an Alfred Deakin Postdoctoral Research Fellowship. VC is supported by a Canadian Institutes of Health Research (CIHR) New Investigator award and a University of Alberta Killam Accelerator Award. JS is supported by a leadership level 2 Fellowship, National Health and Medical Research Council (GNT 1176885).

    • Competing interests None declared.

    • Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.