An important consideration in considering the benefits of more sleep is the compositional nature of time. In the fixed 24hour day allotted to us all, any lengthening of sleep duration must displace an equal and opposite amount of time in all other activities-in movement terms, the combined time spent in sleep, sedentary, light, moderate, and vigorous activities must equal 100% of a 24-hour day. Is it the longer sleep or the reduced time spent in other activities that is beneficial? This is important because the activities displaced by additional sleep may themselves be desirable. For example, in our own population study using compositional analysis, we found that the best mean daily sleep duration for 11-to 12-year-old children ranged widely, from 8.3 to 11.4 hours, depending on the outcome under study. In general, days with highest physical activity and lowest sedentary time were optimal for physical health, while days with highest sleep were optimal for mental health, and days with highest sedentary time and lowest physical activity and sleep were optimal for cognitive and academic outcomes, presumably reflecting more time for schoolwork and studying. 11 However, these data are cross-sectional and observational. If large trials incorporating pre-post accelerometry and 24-hour time diaries were to successfully replicate the findings in this meta-analysis, then they could show which reciprocal time use changes actually cause the best benefits for diverse areas of children's functioning. This could also provide vital experimental evidence about how sleep and other components of time influence metabolic health metrics, such as adiposity, inflammatory markers, satiety markers, and glucose regulation.For now, a simple intervention-shifting bedtime earliercould improve sleep duration for the many children who are chronically sleep deprived. Addressing the social drivers of late bedtimes (homework, social media use) is also key if we are to realize earlier bedtimes in children at scale.