Reading the study, it seems like the statistics are not so clear cut. Link:
"Data from the largest study of its kind in the U.S. reveal a 25 percent jump in the number of heart attacks occurring the Monday after we “spring forward” compared to other Mondays during the year – a trend that remained even after accounting for seasonal variations in these events. But the study showed the opposite effect is also true. Researchers found a 21 percent drop in the number of heart attacks on the Tuesday after returning to standard time in the fall when we gain an hour back."
It's a bit odd for the shift back and forth to be so closely correlated. Later in the study:
"Total daily admissions were adjusted for seasonal and weekday variation, as the rate of heart attacks peaks in the winter and is lowest in the summer and is also greater on Mondays and lower over the weekend."
A quick check in Excel tells me that for the period of the study (January 1, 2010 to September 15,2013), there is one more Friday, Saturday and Sunday than the rest of the week. There is no mention of this fact being adjusted in their results.
Then let's look at what day of the week the year starts:
1/1/2010 = Friday
1/1/2011 = Saturday
1/1/2012 = Sunday
1/1/2013 = Tuesday
Starting to notice something? 2012 was a leap year. No mention for any adjustment for that either. 2/29/12 was a Wednesday, by the way.
So, Mondays are statistically have the highest average heart attacks, most likely because there is one more weekend in the data and that one weekend is most likely in the winter (end of the year) when the number of heart attacks are lower. How much lower? A range of 12-18% lower would account for the 25% versus 21% in heart attack rates between springing forward and falling back.
There was a similar study several years ago that showed a similar correlated link of 10% between the day daylight savings movements as well:
But I could never find the study mentioned. I'm willing to bet there was also an extra winter weekend weekend or two they did not account for in their study as well.
Whenever you see correlations so tidy like that, there has to be something going on with the data.