Retraction: Suicide after Natural Disasters

In the February 5 issue, we reported the results of a study designed to determine whether natural disasters affect suicide rates.1 Previous research on the victims of disasters had reported an increased prevalence of post-traumatic stress disorder and depression,2-4 known risk factors for suicidal thinking.5,6 In our article, we reported increases in suicide rates among the people living in the 377 U.S. counties affected by a single, severe natural disaster between 1982 and 1989.
We regretfully report that we have discovered an error in computer programming and that our previous results are incorrect. The error was discovered when we tried to repeat the analysis to determine whether there are changes in homicide rates after natural disasters. When we retraced the process of assembling the set of data on suicide, we discovered that deaths occurring in 1990 had been counted twice. Because we included only disasters that occurred between 1982 and 1989, 1990 could only be a postdisaster year. The double counting of suicides in 1990 raised the postdisaster rates to levels that differed statistically from the predisaster rates. After the error was corrected, a new analysis showed no significant increase in suicide rates after natural disasters, either for all types of disasters combined or for individual types of disasters.