This interdisciplinary project in population health will determine whether there are secular trends in sleep duration and timing for adults and children and whether economic and demographic factors affect sleep duration. Sleep comprises approximately one-third of a person's life. Historically, sleep duration has received relatively little attention in studies of health. However, recent years have seen mounting evidence that how much people sleep affects diverse health and functional domains, including memory and learning, immune function, carbohydrate metabolism, appetite, and mortality. Both experimental and observational studies suggest that persons who chronically sleep short hours may accelerate the development or increase the severity of age-related health problems such as obesity and diabetes. Considering the consequences of insufficient sleep, a secular decrease in sleep duration would be significant for the health and well-being of the population. A few surveys over the past 40 years have asked about usual sleep hours and there appears to be a trend to shorter sleep hours in recent years. However, this survey question may have weak validity. Thus, controversy exists concerning whether or not sleep duration has in fact declined. The present project aims to overcome the limitations of such data by analyzing a series of datasets primarily used by labor economists that include 24-hour time diaries. Data were collected in adults beginning in 1965 and in children beginning in 1981. The first aim is to determine if time allotted for sleep has declined over the past 40 years for adults or over the past 20 years for children. Furthermore, we will examine whether sociodemographic characteristics such as race, income, workforce participation and family structure predict time allotted for sleep and whether these associations have changed over time. There is also evidence that shiftworkers, who "override" their circadian clocks by sleeping during the day, are at increased risk of accidents, reduced productivity, cardiovascular disease, and gastrointestinal disorders. Therefore, we are also interested in whether there are trends in when people sleep. We will also examine whether the bedtimes of children have changed. The results of these analyses will provide novel insight into historical trends in sleep behavior. Sleep duration may be on the causal pathway between sociodemographic factors and health and may contribute to health disparities.

R01HL082907

2006-03-01

2009-02-28