To our knowledge, this study is the first description of sleep duration and nap frequency in families of those with type 1 diabetes. We found that 41% of the CCDR Family Study participants were sleeping insufficiently. Our results on sleep duration in children and adolescents are similar to NSF polls (National Sleep Foundation, 2004
; National Sleep Foundation, 2006a
), which found that 52% of children aged 5–11 and 80% of adolescents aged 12–19 had insufficient sleep. The proportions with insufficient sleep in the current study were 52% and 64%, respectively. The NSF polls also reported that 18% of school-aged children and 38% of adolescents take regular naps, with the proportions being higher in the current study (33% and 53%, respectively), but not significantly so. Taken together, these results demonstrate that children and adolescents do not get sufficient amounts of sleep. Because adequate sleep is particularly vital as children and adolescents pass through important physical, psychological, and developmental stages, sleep deprivation at a young age may represent a significant health burden. For example, insufficient sleep contributes to excessive daytime sleepiness and school performance problems, and it is associated with depressed mood (Fallone, Owens, & Deane, 2002
). There is also increasing evidence that insufficient sleep is a risk factor for obesity, insulin resistance, and diabetes (Spiegel et al., 1999
; Vorona et al., 2005
Previous studies in adults have found that less sleep increased the risk of developing diabetes and insulin resistance (Ayas et al., 2003
; Gangwisch et al., 2007
; Knutson et al., 2006
; Mallon et al., 2005
). Our analysis showed no association of either diabetes or insulin resistance with sleep duration or insufficient sleep, possibly because the sample size was relatively small. However, it is also possible that, in subjects already at elevated risk of diabetes by virtue of being part of a family with an individual with type 1 diabetes, sleep may not add any additional risk beyond the genetic and lifestyle factors within these families.
The current analysis also did not demonstrate significant associations between sleep duration and obesity, in contrast to other research (Chen et al., 2008
; Knutson et al., 2006
; Taheri, Lin, Austin, Young, & Mignot, 2004
; Vorona et al., 2005
). However, we did find significant associations between regular napping and lower adiposity as measured by BMI, percent body fat, and WC in adolescents.
Research on the relationship between napping and nighttime sleep has been limited, and interpretations based on this research are inconclusive. The results of the current study showed no association of insufficient nighttime sleep with regular napping. Similarly, other research has found that there were no significant differences in nighttime sleep quantity or quality between adults who did and did not nap (Pilcher, Michalowski, & Carrigan, 2001
). While regular daytime napping may not fully compensate for insufficient nighttime sleep, it may help to decrease negative effects such as daytime sleepiness and poor performance (Vgontzas et al., 2007
). Also, as was indicated by our results, regular napping in adolescents may offer specific protective effects against obesity in addition to nighttime sleep. However, further research is needed. The highly significant association of NH Black race/ethnicity with regular naps remained when adjusted for other covariates in all age groups. Others have found similar racial differences in regular nap behavior in children (Crosby, LeBourgeois, & Harsh, 2005
), perhaps reflecting cultural differences in napping acceptability.
Many potential biological mechanisms have been proposed to explain how obesity and diabetes may be associated with decreased sleep duration. One study showed that sleep debt significantly lowers glucose tolerance and increases cortisol in young, healthy adults, indicating that insufficient sleep affects carbohydrate metabolism and endocrine function (Spiegel et al., 1999
). Another study reported that perceived sleep debt (desired hours of sleep minus actual hours of sleep) and shorter weekday sleep duration were significantly associated with higher hemoglobin A1c, indicative of poorer blood sugar control, in adults with diabetes (Knutson et al., 2006
). Reduced levels of leptin, a hormone that regulates hunger and metabolism, have been associated with shortened sleep in adults (Taheri et al., 2004
). Studies in children and adolescents have shown that obese children with shorter sleep duration also have increased insulin resistance and that children with obstructive sleep apnea (OSA) have higher levels of insulin resistance when compared to those without OSA (Flint et al., 2007
). Unfortunately, the current study did not include a measure of OSA. Thus, obesity, diabetes, and sleep may be linked through alterations in energy metabolism, endocrine dysfunction, leptin, and/or OSA.
One limitation of the present study is the use of self-reported sleep. However, other subjective sleep assessments have been found to be valid when compared to actigraphic sleep measurements (Lockley, Skene, & Arendt, 1999
). The measurement of sleep duration over 1 day may also not accurately reflect usual sleep patterns, though follow-up questions were asked to determine if the participants’ reported sleep was typical. Furthermore, the interval between reported sleep and wake times is a relatively crude measure for sleep duration, as it more correctly estimates time in bed which could include time spent trying to fall asleep (sleep latency). Unfortunately, we did not collect information on OSA and nap duration; therefore, total sleep duration over a 24-hr period cannot be assessed. The fact that many of the children’s questionnaires were completed by their parents might have led to some inaccuracies, although good agreement has been found between child- and parent-completed questions on sleeping problems and tiredness (Sundblad, Saartok, & Engström, 2006
). Finally, using the age-specific AASM recommendations (Lamb, 2006
; National Institute of Neurological Disorders & Stroke, 2006
) for categorizing adequate vs. insufficient sleep duration is also a limitation because sleep needs may vary from person to person. In future studies, examining perceived sleep debt may be a way to address this variation.
The use of a cross-sectional study design precludes conclusions about the direction of the association between insufficient sleep and diabetes. Previous research supports an increased risk of diabetes in individuals with reduced sleep, but diabetes may also alter sleep. For example, almost 50% of children and adolescents with type 1 diabetes experience nocturnal hypoglycemia (Beregszaszi et al., 1997
). Young individuals with type 1 diabetes also experience more awakenings during sleep from rapid changes in glucose levels (Pillar et al., 2003
). Therefore, diabetes may affect, not only the sleep of the individual with diabetes, but also the sleep of their families due to frequent nightly arousals and increased anxiety over nocturnal hypoglycemia.