"The fact that it's such a polar opposite shift is really surprising," said lead author Matthew Peters at Tulane University in New Orleans.

After the 2005 storm, the overall number of attacks tripled, likely due to an increased number of smokers. Attacks were more likely on weekday evenings and weekends, according to data published in the American Journal of Cardiology.

Before the hurricane, 23 percent of heart attacks happened on Mondays, in line with national averages. This fell to 10 percent after the storm, less than on any other day of the week.

Previous research has attributed the usual excess of Monday and morning heart attacks to work stress, which also peaks on Monday and in the morning, but researchers were puzzled to find the pattern reversed after Katrina.

"Studies before have looked at the overall increase in (heart attack) after disasters, like the earthquake in Japan and Katrina, but this is the first time we've looked at the timing," Peters said.

If the result holds up after further study, the authors offer a tentative explanation: sources of stress change in a city following a major disaster.

Peters suggests that for people put out of work by the storm, Mondays stopped being especially draining. For those still working, a job became a welcome break from stresses of hurricane related fallout at home.

"People aren't getting up and going to work, and home lives are even more difficult," Peters said.

But the idea of "stress relief at work" doesn't account for all the findings, said David Krantz, who studies stress induced heart attacks at Uniformed Services University of the Health Sciences in Bethesda, Maryland.

Morning heart attacks have been attributed to hormone shifts and increased strain on the heart while waking up, which wouldn't be affected by a hurricane, said Krantz.

Data were limited to one medical center, Tulane University Hospital. A previous study based on data from the same hospital found that after the storm, more heart attack patients lacked insurance, had temporary housing, were divorced and failed to regularly take prescribed medications.

"(The study) doesn't confirm yet that experiencing the storm changed people, the storm may simply have changed who remained in town," said Kathryn Roecklein, who studies the influence of body rhythms on biology at the University of Pittsburgh.

Katrina displaced 1.3 million Gulf Coast households, who were dispersed through all 50 states, temporarily or permanently.

"A disaster that doesn't cause people to move away from the area might make for a better test of this question," she added.

The study also did not take into account other factors that changed. Post-Katrina patients were on average younger, more unemployed, less likely to be insured and more likely to smoke, which is a sizeable risk factor for heart attack. These changes might have influenced the shift in heart attack timing, but Krantz said it's difficult to tell. SOURCE: bit.ly/118Sc10

(Reporting from New York by Kathryn Doyle at Reuters Health; editing by Elaine Lies)

• A Permanent Economic Emergency