Lauderdale et al. Respond to 'How Much Do We Really Sleep?'

摘要

We enjoyed reading Stuart Quan's wide-ranging and poetic perspective on the importance of sleep, and we thank him for his insightful commentary (1) on our paper (2). Noting the unexpectedly low average sleep duration in our study, he touched upon two possible sources of bias; both are intriguing suggestions. The first is that actigraphy might underestimate sleep compared with other measurement modes, and the second is that this cohort, based in Chicago, Illinois, is atypically sleep deprived.Validation studies have concurrently measured sleep with polysomnography and actigraphy in a sleep laboratory setting, where polysomnography was the "gold standard" (3-6). These studies found no evidence that actigraphs consistently over- or underestimate sleep duration and latency in healthy subjects. However, we can speculate that there might be a fundamental problem with such validation studies if people sleep differently when attached to multiple sensors during polysomnographic monitoring, perhaps moving less than they normally do during sleep. Were that the case, then actigraphy in the home梬here sleepers may be more active梞ight systematically underestimate sleep compared with actigraphy in the monitored sleep laboratory subject. There is no obvious way to prove or disprove this possibility