Daylight Savings Time Switch Carries Stroke Risk

Making the switch to daylight savings time can send some people to the hospital, a study from Finland shows.

For many people the extra hour of daylight that will come starting Sunday with Daylight Savings Time will be energizing.

But for others, the time change may be dangerous--at least for a while.

Moving the clock ahead brings an increased risk of having a stroke, researchers in Finland found.

Moving it back when time changes in the fall has a similar risk.

"Previous studies have shown that disruptions in a person's circadian rhythm, also called an internal body clock, increase the risk of ischemic stroke, so we wanted to find out if daylight saving time was putting people at risk," said Jori Ruuskanen MD, Phd, (photo right) lead author of a study due to be presented American Academy of Neurology's annual meeting in Vancouver, Canada on April 19.

Ischemic stroke is the most common type, he said.

Ruuskanen and his co-authors looked at stroke data from Finnish hospital registries for patients hospitalized from 2004 to 2013. They found an 8% increase in the rate of stroke for the 3,033 people hospitalized during the week after daylight savings time compared with the 11,801 people hospitalized in the two weeks before and two weeks after the time change.

In an email interview with MD Magazine, Ruuskanen, a neuro-hospitalist at Turku University Hospital in Turku, Finland, answered questions about how ischemic stroke and daylight saving time seemed to be related

What reason have previous studies given for a change in circadian rhythm increasing the risk of stroke. In other words how does the one-hour change cause the increased risk?

Stroke risk is highest in the morning hours and we know from previous studies that daylight savings time (DST) change slightly shifts the timing pattern of stroke onset. Previous studies have also shown that the disruption of the circadian clock due to other reasons—such as rotating shift work--and sleep fragmentation are associated with an increased risk of stroke. However, we did not know whether stroke risk is affected by DST transitions. What is common in these situations is the disturbed sleep cycle, while the immediate mechanisms for the increased risk remain unknown at the moment.

Why is the impact of the time change temporary? The stroke rate was no higher after the first two days following the time switch.

The body, and the circadian clock, will adapt to the change. In previous studies on myocardial infarction and DST change the effects have been visible during the [entire] week following the switch.

There were more strokes, but not more deaths? Is there any possible explanation for that?

It is probable that some strokes occurred earlier than they otherwise would have. We reported hospital mortality rate, not absolute numbers, a rate which was approximately 3% in the total study population. Daylight saving time transition did not affect in-hospital mortality rate, and I would not have expected to see a big change in mortality rate per se.

People with cancer were 25% more likely to have an ischemic stroke DST, while the overall rate for ischemic stroke increased by 8 percent. What is there about cancer that could make that difference?

Older age and cancer are well known risk factors for stroke. These people may be more prone to sleep difficulties, but there may also be other factors, such as changes in immunology related to cancer, which interplay with sleep disturbance to increase the risk further.

Could there be a similar risk with the switch back to standard time in the fall?

The change in [stroke] incidence was similar with switch forward and back.

What are the limitations of your study?

As we compared weeks after DST shift to similar weeks without shifts over a decade it is hard to see any specific factors other than the shift that could be identified to differ between these weeks and affect stroke incidence. This is a registry study and the limitations are related to this (eg, misdiagnoses and coding errors are a possibility in a registry). Also, we did not have data on timing of symptom onset but instead had to use time of admission. However, these factors are not likely to differ between the study and control weeks, so I really do not see these as important limitations.

Finland has periods in summer when the days are extremely long and in winter the nights very long, doesn't it?

That’s correct, especially in the northern parts of the country.

Does this make any difference in the impact of a change of one hour? Might it be different elsewhere?

During the time of the DST change in the spring and in the winter the length of the day is nearly the same in the whole country. But it certainly would be interesting to see this kind of study repeated in another country.