Goldilocks Days: optimising children’s time use for health and well-being

Abstract

BackgroundOne size rarely fits all in population health. Differing outcomes may compete for best allocations of time. Among children aged 11–12 years, we aimed to (1) describe optimal 24-hour time use for diverse physical, cognitive/academic and well-being outcomes, (2) pinpoint the ‘Goldilocks Day’ that optimises all outcomes and (3) develop a tool to customise time-use recommendations.MethodsIn 2004, the Longitudinal Study of Australian Children recruited a nationally-representative cohort of 5107 infants with biennial follow-up waves. We used data from the cross-sectional Child Health CheckPoint module (2015–2016, n=1874, 11–12 years, 51% males). Time use was from 7-day 24-hour accelerometry. Outcomes included life satisfaction, psychosocial health, depressive symptoms, emotional problems, non-verbal IQ; vocabulary, academic performance, adiposity, fitness, blood pressure, inflammatory biomarkers, bone strength. Relationships between time use and outcomes were modelled using compositional regression.ResultsOptimal daily durations varied widely for different health outcomes (sleep: 8.3–11.4 hours; sedentary: 7.3–12.2 hours; light physical activity: 1.7–5.1 hours; moderate-to-vigorous physical activity (MVPA): 0.3–2.7 hours, all models p≤0.04). In general, days with highest physical activity (predominantly MVPA) and low sedentary time were optimal for physical health, while days with highest sleep and lowest sedentary time were optimal for mental health. Days with highest sedentary time and lowest physical activity were optimal for cognitive health. The overall Goldilocks Day had 10 hours 21 min sleep, 9 hours 44 min sedentary time, 2 hours 26 min light physical activity and 1 hour 29 min MVPA. Our interactive interface allows personalisation of Goldilocks Days to an individual’s outcome priorities.Conclusion‘Goldilocks Days’ necessitate compromises based on hierarchies of priorities for health, social and economic outcomes.

Evolutionary Time-Use Optimization for Improving Children's Health Outcomes

“…The research described in this paper extends previous work proposed in [16] by considering four decision variables: daily time allocation to sleep, sedentary behaviour, LPA and MVPA, and four health objectives for children: body mass index (BMI), cognition, life satisfaction and fitness. Firstly, we formulate the one-day time-use optimization problem as a single-objective problem in continuous space by optimizing one of the four presented health outcomes.…”

“…Participants were fitted with a wrist-worn accelerometer (GENEActive, Activinsights Ltd, UK) by a trained researcher, with instructions to wear the device 24 hours a day for eight days. Following the return of the device, activity data were downloaded and processed following published procedures [16,19] to determine the average daily minutes spent in sleep, sedentary time, LPA and MVPA. BMI was derived from the child participant's measured height (Invicta 10955 stadiometer) and weight (2-limb Tanita BC-351 or 4-limb InBody 230).…”

“…To address the lack of evidence for optimal time-use allocations, a recent study [16] used compositional linear regression [17] to model the relationship between how children allocated their daily time to four activities (sleep, sedentary behaviour, light physical activity (LPA) and MVPA) and twelve outcomes spanning physical, mental and cognitive health domains. Compositional data analysis enabled all four activities to be included in a single model whilst ensuring their constant-sum constraint to 24 hours was respected [1].…”

Many different roads lead to Rome: equivalence of time-use for activity, sedentary and sleep behaviours and dietary intake profiles among adolescents

“…An iterative approach was used to determine equivalences of time-use behaviours and dietary intake. The multiple linear regression model described above was used as a predictive formula to estimate physical functioning for systematically created time-reallocations and dietary choices [16]. Starting from the sample average (time-use compositional mean rounded to nearest 10 min) an equally spaced grid was created, simulating all possible time reallocations from − 30 to + 30 min in 10-min increments.…”

“…health behaviours (i.e. across the time-use and diet profile), decision tools are needed to ensure optimal understanding of required changes and their benefits for health [16]. Indeed, a significant challenge in health research is ensuring that findings are translated and disseminated in a way that facilitates behaviour change beyond the end of intervention periods (i.e., health behaviours are adopted in real-life settings) [5].…”