Australian study of 10- to 15-year olds shows significant decline in sleep duration between 1985 and 2004

In the current issue of Acta Paediatrica, Dollman et al. (1) present data in support of the notion that there is a secular trend towards declining sleep duration among children and adolescents. The topic of sleep curtailment has a bearing on several aspects of child health, including psychological functioning, and is also possibly linked to childhood obesity.

It is a common view that sleep duration has been declining in Western populations over the past century and that sleep loss has generated a widespread sleep debt. The term ‘sleep debt’ is used to denote cumulative effects of chronic partial sleep curtailment. Later bedtimes, attributed to aspects of modern lifestyle, such as electric light, and now also 24-h availability of television and the Internet, are assumed to be major causes of decline in sleep duration. Yet, there is still a scarcity of systematic data about the trends in the duration of sleep in adults as well as in children and adolescents.

It is particularly important to clarify if there is indeed a significant trend towards shorter sleep in childhood populations. Acute sleep loss has even more negative effects in children than in adults (2), and chronic sleep loss interferes with developmental processes and impairs children's physical functions as well as their mental health (3,4). Also, recent studies indicate that there may be a link between the pandemic increase of childhood obesity and sleep curtailment (5). Nature, significance and magnitude of the associations between sleep and obesity are, though, not yet fully understood.

However, it should be acknowledged that there is excellent research which challenges the perception that children's sleeping times have declined to such an extent that negative health effects have been generated on a mass scale. To illustrate, Iglowstein et al. investigated sleep duration among Swiss children from infancy to adolescence (6). Their study provides reference values regarding sleep duration in children at different ages, and also information about generational trends. They did find that the sleep duration had declined when they compared children in the 1970s and 1990s, but did not interpret their finding to be evidence of a widespread chronic sleep loss in the latter cohort. On the contrary, and in line with the opinions of other sleep researchers, the authors emphasized that the sleep–wake timing is determined not merely by social cues, such as parental pressure, but to a great extent by biological processes. In a scientific debate, the research group hypothesized that a possible cause of the trend towards later bedtimes for children was an increase in positive child-oriented attitudes of parents towards evening bedtime (7). Observations of lower frequency of parent-reported bedtime resistance in their later cohorts supported this hypothesis. To clarify, when children are put to sleep at a time when they are really tired, they are less prone to bedtime struggles and their overall sleep may benefit. The authors therefore concluded that their observations were ‘indicating that parents in the 1990s may have adjusted bedtime to their children's actual sleep need more appropriately than parents in the 1970s’. Yet, the authors cautioned that delaying bedtime too much might lead to sleep restriction with negative effects to the child. Additionally, it has been emphasized that an important task of parents and childcare professionals is to determine optimal amount of sleep for the individual child while taking factors such as age and individual sleep need into account (8).

Thus, sleep is an important aspect of child health, and there are some debates regarding interpretations of existing data on sleep trends and a lack of systematic studies of the secular trends. Therefore, the present study by Dollman et al. (1) is a valuable contribution to the knowledge of the topic. These authors compared the self-reported school-day bedtime and wake-up time in representative samples of 10-to 15-year-old south Australians in 1985 and 2004. Their information was based on the data obtained by using identical questions in the two surveys. The authors found that both the girls and boys of the 2004 sample slept less than their counterparts of the 1985 sample. The decline in sleep duration, which was due largely to later bedtimes, was most pronounced in boys of lower socio-economic status (SES). Moreover, the authors commented that the overall decline in sleep duration of approximately 30 min between 1985 and 2004 coincides with the difference in sleep duration between normal weight and overweight children found in a prospective study of 150 children (9).

On the basis of their findings of declining sleep duration, the authors suggest that there is a need to routinely survey sleep behaviour in youth populations. They point out that the Sleep in America Poll identified that more than 50% of US adolescents report sleeping less than they need. However, many of the US parents, who also participated in this survey, were unaware that their adolescents were not getting sufficient sleep. Dollman et al. propose that such an ‘awareness gap’ between generations can be best addressed if young persons' sleeping habits are regularly profiled by using data from both self and parental report. With regard to the aspect of parent–child interactions, several authors have recently emphasized that the clinical experience as well as neurobiological findings indicate that children need clear adult guidance and support at least until they reach the age of 16 years and probably longer (10). It may be assumed that children's need for adult guidance and support is of particular importance in the area of sleep habits. Several sleep hygienic measures may prove valuable; for example, restricting the time that the youngsters spend in front of TV and computers, adherence to regular bed- and wake-up-times (which should be determined in accordance with the individual child's age and sleep need), exposure to bright light in the morning and encouraging physical activity during the day.

Finally, in their conclusions, Dollman et al. underscore that there is a great need for further research is in the area of paediatric sleep. The authors' finding that ‘lower SES' boys are at especially high risk of chronic sleep loss seems especially important in this respect. It is well known that there is a relationship between SES and several aspects of child health (11,12). The precise nature of the association between lower SES, male gender and a particularly pronounced trend towards declining sleep duration, which was found in the present study, deserves further exploration in future investigations.

The study by Dollman et al. presents results in support of the view that the sleep duration is declining in children and adolescents. These results are a source of concern and should inspire policy-makers and child health planners to take into account the aspect of children's sleep habits into their health programmes.