Pena analyzed Social Security Administration records for 25 million people who died between 1998 and 2011. Births and deaths don't happen evenly across the year, so Pena's analysis controls for seasonal patterns by comparing birthday death rates to what would be expected on a given day.
The Social Security records themselves don't provide information on the cause of deaths or characteristics of the departed. But some patterns do emerge.
For instance, young people between 20-29 years old have an excess death rate of 25.4 percent —perhaps a sign of birthday celebrations gone wrong. Excess death rates are slightly higher for birthdays falling on weekends across all age groups, Pena's the analysis finds, while the excess death rate for 20-29 year olds almost doubles to 48.3 percent.
The study isn't the first to cast a sad shadow on birthday celebrations. A 2012 study in the journal Annals of Epidemiology found a much higher probability (about 14 percent) that people over 60 will die on their birthdays. That study, which analyzed more than 2 million deaths, found that strokes and heart attacks rose 18.6 percent and 21.5 percent, respectively, on birthdays. Tragically, the study also found a 34.9 percent increase in suicides and a 44 increase in deaths from falls.
Pena, whose results will be published in the February 2015 issue of Social Science & Medicine, writes that his study sample is about 10 times larger than the 2012 study. He says further research is needed into what's behind the link between birthdays and deaths, such as the causes and the socioeconomic factors, to drive a potential policy response to prevent these deaths.
"In principle, those policies could be as simple as campaigns to increase awareness of the risks associated with changes in behaviors on or around birthdays," he writes.
In other words, celebrate responsibly.