Health-related quality of life in school-aged boys and girls: longitudinal associations with physical activity and sedentary behaviour in the HAPPY study

Abstract

Background

Health-related quality of life (HRQoL) is positively associated with physical activity (PA) and negatively associated with sedentary time (SED) assessed by self- and proxy-report in children. However, there is a dearth of longitudinal evidence using device-measured PA and SED in this population. This study aimed to examine longitudinal associations of device-measured PA and SED as well as parent-reported screen time with HRQoL in primary (elementary) school-aged boys and girls.

Methods

Data from waves 2 and 3 (child ages 6–8 and 9-11y, respectively) of the Healthy Active Preschool and Primary Years (HAPPY) Study were analysed. At wave 2, children wore ActiGraph GT1M accelerometers to assess time in SED and light-, moderate- and vigorous-intensity PA (LPA, MPA and VPA). Parents reported their child’s usual screen time (television viewing, computer/internet use, computer/electronic games, handheld electronic games). At wave 3, parents reported their child’s HRQoL using the Pediatric Quality of Life Inventory. Mixed-effects linear models were used to assess longitudinal associations of different intensities of PA, SED, and screen time with HRQoL scores. Analyses were stratified by sex of the child and adjusted for child age, BMI z-score and maternal education, with child and preschool/childcare centre of recruitment as random-effects.

Results

Boys had less SED and more MPA, VPA, TPA, electronic game use, handheld electronic game use and total screen time than girls (all p < 0.05); there were no sex differences in LPA, television viewing, computer/internet use, or HRQoL scores. For boys, VPA predicted better physical HRQoL (beta [β] = 0.18, 95%CI 0.02, 0.35). For girls, using electronic games predicted lower psychosocial HRQoL (β=-0.09, 95%CI -0.16, -0.01), while handheld electronic games and total screen time predicted lower physical HRQoL (β=-0.19, 95%CI -0.33, -0.06, and β=-0.04, 95%CI -0.07, -0.004, respectively) and total HRQoL (β=-0.12, 95%CI -0.23, -0.01, and β=-0.03, 95%CI -0.05, -0.001, respectively).

Conclusions

Higher VPA predicted better physical HRQoL for boys; since boys were more active than girls, girls may not have engaged in sufficient VPA to benefit their HRQoL. Further research is needed to examine specific types of screen use, given the observed detrimental associations with HRQoL for girls.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12887-026-06961-w.

Background

Health-related quality of life (HRQoL) is a multidimensional concept that describes the physical, social, and psychological aspects of well-being and functioning [1]. As such, HRQoL is a good indicator of the overall health of an individual [2] and can distinguish between healthy and clinical populations. HRQoL is often used to monitor population health and impacts on later health and developmental outcomes. The importance of promoting physical activity and minimising sedentary behaviour (i.e., activities undertaken in a sitting or reclining posture and requiring little energy expenditure) for HRQoL has been well documented in adults [3], but less is known about those associations in school-aged children.

A systematic review of healthy children and adolescents (3–18 years) identified 31 studies and found that, in general, physical activity is positively associated with HRQoL, whereas sedentary behaviour is inversely associated with HRQoL, in both cross-sectional and longitudinal studies [4]. However, only six of the 31 included studies were longitudinal, with follow-ups ranging from two to 22 years. Findings from these longitudinal studies were consistent with the overall review evidence.

Another important evidence gap is the lack of studies examining device-measured physical activity and sedentary behaviour (i.e., sedentary time) which may provide a more accurate measure of these behaviours. In Wu and colleagues’ review [4], all but one study relied on self- or parental-report. The only study to use device-based measures [5] was cross-sectional in nature and found that moderate- to vigorous-intensity physical activity (MVPA; usually defined as a brisk walking pace or more intense activity) was positively associated, and sedentary time inversely associated, with psychosocial health and overall HRQoL in unadjusted analyses. However, none of the associations remained once adjusted for sex and body mass index (BMI) [5]. Additionally, that study did not examine light-intensity physical activity (LPA), which is important given emerging evidence suggesting that LPA (such as incidental activity or slow walking) may be important for health [6, 7]. Whilst LPA research with children is limited, studies that have analysed the relationship between LPA and health indicators have found positive associations between LPA and reduced blood pressure, insulin resistance and HDL cholesterol [8].

A further systematic review and meta-analysis of healthy children and adolescents (< 18 years) evaluated the effect of school and childcare based physical activity interventions on HRQoL [9]. Of the 17 studies identified (three of which were also included by Wu and colleagues [4], 13 included device-assessment of physical activity as either the sole measurement of physical activity, or in conjunction with parent-report [9]. Of the 13 interventions, seven interventions reported no change to participants’ physical activity levels, five reported a significant increase to physical activity, and two interventions did not report follow-up physical activity levels [9]. Pooled estimated effect size values indicated small but significant improvements to total HRQoL and subdomains, with no effect difference according to the success of the intervention [9]. Thus, whilst physical activity dose is important for physical health, the psychological and psychosocial benefits from physical activity participation may be key contributors to HRQoL.

Finally, studies investigating the association between sedentary behaviours and HRQoL have focused predominantly on total screen time, with few studies examining individual screen behaviours separately [4]. This differentiation is important given that screen time comes from many different sources [10] and different types of screen time may have different associations with outcomes [11–13]. Nonetheless, almost all of the cross-sectional studies identified by Wu et al.⁴ reported significant associations between sedentary behaviour and poor HRQoL, both overall, and across several domains. Further, all four longitudinal studies reported that increased time spent in sedentary activities was associated with poorer HRQoL [4]. The aim of this study was to determine longitudinal associations of different intensities of device-measured physical activity and sedentary time, and different types of parent-reported screen behaviours, with HRQoL in school-aged boys and girls.

Methods

Participants

Data were drawn from waves 2 and 3 of the Healthy Active Preschool and Primary Years (HAPPY) Study. Initial recruitment (wave 1) occurred through preschools and childcare centres in Melbourne, Australia in August-December 2008 and June-November 2009. All parents/guardians with a child aged 3–5 years attending consenting centres were invited to participate, with 1002 providing consent at baseline [14].

The HAPPY Study was initially designed as a cross-sectional study; however, 77% of parents provided consent to be re-contacted for future studies. These parents were invited three- and six-years post-baseline (August 2011-March 2012 and June 2012-April 2013, children aged 6–8 years; and July 2014-May 2015 and June 2015-May 2016, children aged 9–11 years) to participate in waves 2 and 3, respectively. A total of 567 parents (74% of those who agreed to follow-up) consented to wave 2 and 568 (76%) parents consented to wave 3 (see Fig. 1). For waves 2 and 3, child measures were taken either at home or school (depending on the parents’ preference), with data collected during school terms for both waves. Parents provided written informed consent for their child, and children provided verbal assent to participate in the study and wear an accelerometer at each time point. The Deakin University Human Research Ethics Committee (EC291-2007), the Victorian Department of Education and Early Childhood Development (2011_001008) and the Victorian Catholic Education Office (GE11/0009) approved the study.

Fig. 1.

Participant flow chart

Dependent variables

HRQoL data were collected at wave 3 using the parent version of the Pediatric Quality of Life Inventory (PedsQL™ 4.0; Children’s Hospital and Health Center, San Diego, California) generic core scale. Parent-report measures were utilised due to time and budget constraints limiting the extent of data that could be collected directly from children. The PedsQL™ has established reliability and validity, and comprises 23 items encompassing physical functioning (8 items), emotional functioning (5 items), social functioning (5 items), and school functioning (5 items) [15]. Each item is answered on a five-point Likert scale (0 = never a problem; 4 = almost always a problem), with items reverse-scored and linearly transformed (0 = 100, 1 = 75, 2 = 50, 3 = 25, 4 = 0) so that a higher score indicates better HRQoL. A total scale score (derived from the mean of all 23 items), a physical health summary score (derived from the mean of the 8 physical functioning items), and a psychosocial health summary score (derived from the mean of the 15 items across the emotional, social, and school functioning items) are calculated to provide a summary of HRQoL. The total, physical health, and psychosocial health summary scores were used as the outcome variables, as per previous studies [4].

Independent variables

At wave 2, children were fitted with ActiGraph GT1M uniaxial accelerometers (Pensacola, FL, USA) on an elastic belt at the right iliac crest and instructed to wear them during waking hours (except for during water-based activities, e.g., bathing, swimming) for eight consecutive days. Data were collected in 15-second epochs [16, 17] and were processed using a customised Microsoft Excel macro. Non-wear time was determined as ≥ 20 minutes of consecutive zero counts[16, 18]. Children were required to have data recorded for ≥ 8 h/day on ≥ 4 days (including ≥ 1 weekend day) [19] to be included in the analyses. Sedentary time was defined as ≤ 25, LPA as 26–573, moderate-intensity physical activity (MPA) as 574–1002, and vigorous-intensity physical activity (VPA) as ≥ 1003 counts 15s^− 1 [20]. Total physical activity (TPA) was calculated as the sum of LPA, MPA and VPA. Accelerometer variables (min/day) were adjusted for wear time by standardising using the residuals obtained when regressing on wear time [21].

Parents reported their child’s usual hours and minutes in four leisure-time screen behaviours: television viewing, computer and internet use (excluding games), computer and electronic-game use, and handheld electronic game use during the week (i.e., from Monday to Friday) and on weekends (i.e., Saturday and Sunday). Total minutes during the week and on weekends in each individual screen behaviour were summed and divided by seven to give average daily minutes in each behaviour. Average daily screen time was calculated by summing the daily minutes in each of the behaviours.

Covariates

At wave 1, parents completed surveys and reported their child’s sex and date of birth (used to calculate age at wave 2). At wave 2, parents reported their highest level of education, collapsed into: year 10 or equivalent; year 12, trade, certificate, apprenticeship, or diploma; and tertiary education. Children’s height (m) and weight (kg) were measured by trained researchers at wave 2 using standardised measurement procedures with a Wedderburn Seca portable rigid stadiometer and Wedderburn Tanita portable digital scales respectively [22, 23]. Child BMI was calculated as weight (kg) divided by height (m²) and converted to World Health Organization (WHO) BMI-for-age z-scores [24] for comparability across the age range of the sample.

Data analysis

All analyses were performed in Stata 16.0 (StataCorp, Texas, USA). Descriptive statistics were used to summarise demographic characteristics, and unpaired t-tests and chi-square tests were used to examine differences between boys and girls. Mixed-effects linear models were used to examine longitudinal associations of average daily minutes in different intensities of physical activity (i.e., LPA, MPA, VPA, and TPA), sedentary time (SED), and screen behaviours (i.e., television viewing, computer/internet use, computer/electronic game use, handheld electronic game use, individually, and total screen time) at wave 2 with HRQoL (PedsQL total scale score, physical health summary score, and psychosocial health summary score) at wave 3. While physical activity and screen time were also assessed at wave 3, these variables were not included in the present analyses as the study was designed to examine prospective associations between exposures at ages 6–8 years and outcomes at ages 9–11 years, rather than changes in behaviours over time. Each physical activity and sedentary behaviour variable was entered into the regression models individually (i.e., not mutually adjusted), to estimate its independent association with HRQoL and to minimise potential multicollinearity between related behavioural measures. Models were stratified by child sex and adjusted for child age, BMI z-score and maternal education at wave 2, with child and preschool/childcare centre of recruitment as random-effects. Complete case analyses were conducted for accelerometry outcomes (n = 404) and screen time outcomes (n = 446).

Results

Sample characteristics are shown in Table 1. At wave 2, children (45% girls) had a mean age of 7.6 years. Boys spent less time in SED, more time in MPA, VPA, TPA and total screen time, and more time using electronic games and handheld electronic games than girls (all p < 0.05); there were no sex differences in LPA, television viewing, or computer/internet use. Additionally, there were no sex differences in HRQoL scores at wave 3 (Table 1).

Table 1.

Participant characteristics, mean (SD) unless otherwise noted

	Boys (n=246)	Girls (n=200)	p-value*
Demographic characteristics
Child age (years)	7.6 (0.7)	7.5 (0.7)	0.5
BMI z-score	0.4 (0.9)	0.4 (0.9)	0.4
Maternal education, %			0.1
Year 10 or equivalent	6.9	4.5
Year 12/trade/diploma	24.9	32.0
University degree/post-graduate	68.2	63.5
Parent-reported behaviours (min/day)
Television viewing	82.3 (52.8)	71.9 (48.9)	0.03
Computer/internet use (excl. games)	13.6 (19.4)	14.7 (26.5)	0.6
Computer/electronic game use	16.0 (24.0)	7.7 (24.6)	<0.01
Handheld electronic game use	12.6 (23.4)	8.8 (15.8)	0.05
Total screen time	124.4 (74.0)	103.2 (68.5)	<0.01
Device-measured behaviours (min/day)	(n=227)	(n=177)
SED	374.1 (44.7)	388.8 (42.6)	<0.01
LPA	272.4 (34.0)	270.9 (34.3)	0.7
MPA	47.7 (12.3)	38.5 (9.9)	<0.01
VPA	21.5 (10.4)	17.1 (8.0)	<0.01
TPA	341.6 (44.6)	326.6 (42.3)	<0.01
HRQoL †
Physical HRQoL summary score	62.4 (14.0)	60.5 (15.6)	0.2
Psychosocial HRQoL summary score	54.2 (12.9)	55.7 (13.5)	0.2
Total HRQoL score	56.2 (11.9)	56.9 (12.5)	0.5

Abbreviations: BMI body mass index, HRQoL health-related quality of life, LPA light-intensity physical activity, MPA moderate-intensity physical activity, SED sedentary time, TPA total physical activity, VPA vigorous-intensity physical activity

* p-value for difference between boys and girls; †HRQoL scores from wave 3 (all other data from wave 2)

Table 2 shows adjusted longitudinal associations of physical activity, sedentary time and screen time (wave 2) with HRQoL (wave 3) (unadjusted results are presented in Supplementary Table 1). In adjusted analyses, for boys, VPA predicted better physical HRQoL. For girls, using electronic games predicted lower physical HRQoL and total HRQol. No other associations were identified.

Table 2.

Longitudinal associations^* of physical activity and sedentary behaviour with HRQoL

Minutes/day	Boys			Girls
	Physical HRQoL subscore β (95% CI)	Psychosocial HRQoL subscore β (95% CI)	Total HRQoL score β (95% CI)	Physical HRQoL subscore β (95% CI)	Psychosocial HRQoL subscore β (95% CI)	Total HRQoL score β (95% CI)
Parent-reported behaviours
Television viewing	-0.001 (-0.04, 0.03)	-0.01 (-0.04, 0.02)	-0.01 (-0.04, 0.02)	-0.03 (-0.08, 0.01)	0.004 (-0.04, 0.04)	-0.01 (-0.04, 0.03)
Computer/internet use (excl. games)	0.01 (-0.08, 0.10)	-0.01 (-0.09, 0.07)	-0.01 (-0.08, 0.07)	-0.06 (-0.14, 0.02)	-0.05 (-0.12, 0.02)	-0.05 (-0.12, 0.01)
Computer/e-game use	-0.04 (-0.11, 0.04)	-0.03 (-0.10, 0.04)	-0.03 (-0.09, 0.03)	0.01 (-0.08, 0.10)	-0.09 (-0.16, -0.01)	-0.06 (-0.13, 0.01)
Handheld e-game use	-0.01 (-0.09, 0.06)	0.004 (-0.06, 0.07)	-0.001 (-0.06, 0.06)	-0.19 (-0.33, -0.06)	-0.09 (-0.21, 0.03)	-0.12 (-0.23, -0.01)
Total screen time	-0.01 (-0.03, 0.02)	-0.01 (-0.03, 0.01)	-0.01 (-0.03, 0.01)	-0.04 (-0.07, -0.004)	-0.02 (-0.05, 0.005)	-0.03 (-0.05, -0.001)
Device-measured behaviours
SED	-0.02 (-0.06, 0.02)	0.01 (-0.02, 0.05)	0.01 (-0.03, 0.04)	0.005 (-0.05, 0.06)	0.01 (-0.04, 0.06)	0.01 (-0.04, 0.05)
LPA	-0.004 (-0.06, 0.05)	-0.03 (-0.08, 0.02)	-0.02 (-0.07, 0.02)	-0.03 (-0.10, 0.04)	-0.01 (-0.06, 0.05)	-0.01 (-0.07, 0.04)
MPA	0.10 (-0.04, 0.24)	-0.05 (-0.19, 0.09)	-0.01 (-0.14, 0.12)	0.14 (-0.10, 0.38)	-0.03 (-0.22, 0.17)	0.01 (-0.17, 0.20)
VPA	0.18 (0.02, 0.35)	0.06 (-0.10, 0.22)	0.09 (-0.05, 0.24)	0.22 (-0.07, 0.51)	-0.04 (-0.28, 0.20)	0.03 (-0.20, 0.26)
TPA	0.02 (-0.02, 0.06)	-0.02 (-0.06, 0.02)	-0.01 (-0.04, 0.03)	-0.01 (-0.06, 0.05)	-0.01 (-0.05, 0.04)	-0.01 (-0.05, 0.04)

Abbreviations: CI confidence interval, e-game – electronic game, HRQoL health-related quality of life, LPA light-intensity physical activity, MPA moderate-intensity physical activity, SED sedentary time; TPA total physical activity, VPA vigorous-intensity physical activity

^* Adjusting for child age, zBMI and maternal education at wave 2 and clustering by centre of recruitment; β = beta co-efficient. Bold font indicates p < 0.05

Discussion

This study aimed to determine longitudinal associations of device-measured physical activity and sedentary behaviour and parent-reported screen behaviours with HRQoL in a sample of Australian children from ages 6–8 to 9–11 years. We did not find consistent associations for boys and girls. For boys, we found just one association, with VPA predicting better physical HRQoL. Whereas, for girls, computer and electronic game use (including handheld electronic games) and total screen time predicted poorer physical, psychosocial and total HRQoL.

In our study, we found that higher levels of VPA at age 6–8 years predicted better physical HRQoL at age 9–11 years for boys only. Of the few previous longitudinal studies that have investigated associations of physical activity with children and adolescents’ HRQoL, the majority report significant longitudinal or temporal associations between these variables [25–30]. However, it is important to note that none of these studies stratified analyses by sex of the child; just one study [25] examined whether the relationship between sports participation and HRQoL differed by sex, finding that girls who were classified as sports ‘non-participants’ or ‘dropouts’ had significantly lower HRQoL at 2-year follow up compared to boys. Additionally, all these previous studies relied on proxy- or self-report physical activity. One study has examined compliance with 24-hour movement guidelines and reported no association between meeting the physical activity component of the guidelines (measured by accelerometry) at ages 3–5 years and HRQoL at ages 9–11 years [31]. However, as the authors noted, the guidelines are based on limited evidence and may not represent the threshold of physical activity beneficial for HRQoL, which may explain why we observed an association when examining continuous time in physical activity rather than dichotomising the data according to the guidelines. In addition, this null association may have been due to guidelines for children aged 3–5 years providing recommendations for total physical activity (i.e., 180 minutes of any intensity), but our findings suggest that only VPA is associated with HRQoL, and for boys only. Given the significant difference in volume of VPA we observed between boys and girls (almost 5 minutes per day), it may be that girls are not engaging in sufficient amounts of VPA to confer benefits to their HRQoL.

We found that higher total screen time predicted poorer physical and total HRQoL scores for girls only. In addition, when examining individual screen behaviours, we found that computer/electronic game use predicted poorer psychosocial HRQoL, while handheld electronic game use predicted poor physical and total HRQoL for girls. The absence of this finding in boys despite their higher screen use in girls, may be due to the mediating effects from time spent in VPA. Although a number of studies support negative associations between total screen time and HRQoL (e.g., [32, 33]) few examined individual screen behaviours. One previous longitudinal study found that higher amounts of television viewing were associated with lower overall and psychosocial HRQoL in 10- to 13-year-old girls but, contrary to our findings, that higher videogame use was associated with lower HRQoL across all domains in boys only [34]. In that study the median videogame use among girls was < 0.01 min per day, indicating that more than half of girls reported no engagement with videogames. Given their interquartile range (22.5), the skewed distribution may have limited the ability to detect associations. In contrast, the use of mean minutes in this study may better capture variation among girls who do engage with videogames, thus the different findings in this study. A study in adolescents (mean age 16 years) reported that high video game use was associated with poorer HRQoL for both sexes [35]. Whilst that study did not assess the types of videogames the adolescents engaged with, it has been suggested that the association between electronic game use and lower HRQoL may be due to increased exposure to violence, therefore triggering anti-social behaviour, fear and anxiety, and decreased perceptions of physical health [36, 37]. However, it is unlikely that the girls in our study (aged 6–8 years) would play games that contain violence. It may be that at this age, screen time (especially sedentary gaming) displaces time spent in physical activity, specifically VPA, which may be more beneficial for HRQoL. The observed sex differences in our study are unlikely to be explained solely by small differences in total MVPA or screen time relative to current recommendations. Instead, they may reflect differences in the intensity, type, or social context of these behaviours, which are not fully captured by aggregate measures. However, these mechanisms remain speculative and warrant further investigation. Regardless, our longitudinal findings suggest that strategies to reduce computer and electronic game use may be needed from as young as 6–8 years of age, and particularly for girls. Whilst the effect sizes for associations in this study appear small, it is important to remember that participants were primary aged children. Longitudinal data with adults in Germany has shown a linear decline in physical health related HRQoL with age [38]. Additionally, non-linear fluctuations in mental health related HRQoL was identified, most likely in response to the significant life events that occur across the lifespan [38]. Physical activity levels are on the decline in developed nations [39], and physical activity levels have been found to decline across the lifespan also, dropping sharply during the transition from childhood to adolescence, when children enter secondary school [40]. Further, a significant decline in VPA has been found to occur in the transition between adolescence and adulthood also [41]. Given the positive association found between VPA and boys’ physical HRQoL in this study, strategies and interventions to increase VPA in childhood, and support girls to replace time spent engaging with video games, may offer a buffer against age-related physical activity declines, and support higher HRQoL.

Strengths and limitations of the current study must be acknowledged. The longitudinal design and the inclusion of device-measured physical activity and sedentary time are key strengths; to our knowledge, this is one of few studies to examine longitudinal associations of continuous physical activity and sedentary time assessed by accelerometry with HRQoL in school-aged children. Limitations of this study include the loss to follow-up and parent reported HRQoL measures; child self-report of HRQoL may have yielded different results. In addition, HRQoL data were not collected for the full sample at wave 2 (these measures were introduced part way through wave 2, with data available for only approximately half the sample); as such, we were unable to control for HRQoL at wave 2 or examine changes in HRQoL over time. Restricting analyses to participants with HRQoL measured at both waves would have substantially reduced the sample size and may have introduced selection bias. Consistent with the study aim of examining prospective associations, HRQoL at wave 3 was therefore treated as the primary outcome. Finally, data on children’s use of ‘newer’ technologies such as smartphones and tablets were not collected as wave 2 data were collected in 2011-13, prior to these devices becoming ubiquitous.

Conclusions

Findings from this study suggest that physical activity, specifically higher intensities of physical activity, are important for boys’ HRQoL, while total screen time and computer/electronic game use are detrimental to girls’ HRQoL. Further investigation to determine the potential mechanisms through which physical activity, sedentary behaviour and screen time affect children’s HRQoL is warranted. In the meantime, lifestyle interventions to improve HRQoL should focus on promoting VPA and reducing screen time. Both parents and clinicians may need to be mindful of these specific physical activity and screen behaviours in children and focus on encouraging higher intensity physical activity for boys and reducing certain screen behaviours for girls in particular.

Abbreviations

BMI

Body Mass Index

HRQoL

Health-related Quality of Life

LPA

Light-intensity Physical Activity

MPA

Moderate-intensity Physical Activity

MVPA

Moderate- to Vigorous-intensity Physical Activity

PA

Physical Activity

SED

Sedentary Time

TPA

Total Physical Activity

VPA

Vigorous-intensity Physical Activity

PedsQL™

Pediatric Quality of Life Inventory

Authors’ contributions

KLD conceptualized and designed the current study, carried out the analyses, drafted the initial manuscript, and reviewed and revised the manuscript. TH, JS, and KDH conceptualized and designed the overall study, provided input into the data collection instruments, and critically reviewed the manuscript for important intellectual content. BdPC, LA, and AT critically reviewed the manuscript for important intellectual content. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Funding

HAPPY was funded by Deakin University and Australian Research Council Discovery Grants (DPI10101434 and DP140100554). KLD was supported by an Alfred Deakin Postdoctoral Research Fellowship during the analyses and write-up of this article. LA is supported by an Australian Research Council Discovery Early Career Researcher Award (DE220100847). TH was supported by a National Health and Medical Research Council Early Career Fellowship (APP1070571) during data collection for this study. JS is supported by a National Health and Medical Research Council Leadership Level 2 Fellowship (GNT1176885). KDH is supported by a Heart Foundation Future Leader Fellowship (105929).

Data availability

The datasets analysed for the current study are not publicly available due to ethical restrictions related to the consent given by participants at the time of study commencement. An ethically compliant dataset may be made available by the corresponding author on reasonable request and upon approval by the Deakin University Human Research Ethics Committee.

Declarations

Ethics approval and consent to participate

Approval to conduct the HAPPY Study was granted by Deakin University’s Human Research Ethics Committee (EC291-2007), Department of Education and Early Childhood Development (2011_001008), and Catholic Education Melbourne (GE11/0009). The study was conducted in accordance with the Declaration of Helsinki. Participating parents provided written informed consent for themselves and their children to take part.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.