On 2014 Mar 11, Daniel Kripke commented:

These wonderful data add important additions to our understanding of long and short sleep. There are some additional issues to which the data might contribute.

1) It appears that like the Cancer Prevention Studies and many others, this study suggests that the longest survival was associated with sleep durations slightly below the mean. According to Table 5, among females, the risk was consistently just a bit less with -1 hour adjusted sleep than with mean sleep. (Incidentally, perhaps the squared age-adjusted sleep regressions for females would be significant if centered at the minimum risk.) Among males, Table 5 indicated that death risk rose much more rapidly above the mean age-adjusted sleep duration than below it. Long sleep was associated with more risk than short sleep. Would the lowest risk for males be associated with sleep durations about 30 min. below the mean, as was the case in the adult data?

2) Increased mortality risk has been found associated with delayed sleep phase disorder, that is, with late bedtimes. Is a late bedtime a mortality risk factor in data from these children?

3) Several studies of risk factors among adults have indicated that use of hypnotic drugs (sleeping pills) is an independent risk factor for increased mortality, though sometimes confounded with short sleep or insomnia. Does this remarkable data set offer any information about mortality risks associated with hypnotics?

On 2014 Mar 11, Daniel Kripke commented:

These wonderful data add important additions to our understanding of long and short sleep. There are some additional issues to which the data might contribute.

1) It appears that like the Cancer Prevention Studies and many others, this study suggests that the longest survival was associated with sleep durations slightly below the mean. According to Table 5, among females, the risk was consistently just a bit less with -1 hour adjusted sleep than with mean sleep. (Incidentally, perhaps the squared age-adjusted sleep regressions for females would be significant if centered at the minimum risk.) Among males, Table 5 indicated that death risk rose much more rapidly above the mean age-adjusted sleep duration than below it. Long sleep was associated with more risk than short sleep. Would the lowest risk for males be associated with sleep durations about 30 min. below the mean, as was the case in the adult data?

2) Increased mortality risk has been found associated with delayed sleep phase disorder, that is, with late bedtimes. Is a late bedtime a mortality risk factor in data from these children?

3) Several studies of risk factors among adults have indicated that use of hypnotic drugs (sleeping pills) is an independent risk factor for increased mortality, though sometimes confounded with short sleep or insomnia. Does this remarkable data set offer any information about mortality risks associated with hypnotics?