On 6/24/19 4:15 PM, Tim Parenti wrote:
I would think, in that case, the magnitude of those effects on health data would be sufficiently small that they would likely be dwarfed by the broader public health trends of the time.
One way to tease out the effects of daylight saving time is to look at the health records of people near a time zone boundary, as you get a one-hour civil-time difference for free. As far as I know this sort of work was pioneered in 2011 by Mikhail Borisenkov of the Komi Science Center, who found that in Russia position within a time zone explained 15% of the variability in female breast cancer mortality. Similar results have been found in the US, as I mentioned in 2017 on this mailing list <https://mm.icann.org/pipermail/tz/2017-November/025548.html>. More recently, researchers have used this technique to find that a one-hour time difference suppresses voting by 2-5%. See: Holbein JB, Shafer JP, Dickinson DL. Insufficient sleep reduces voting and other prosocial behaviors. 2019. Nat Hum Behav 3, 492-500. https://doi.org/10.1038/s41562-019-0543-4 So it appears that daylight saving time causes cancer and keeps voters away from the polls, in addition to its other effects.