An independent review of the data from a case-control study of gun ownership as a risk factor for homicide in the home, reported by my colleagues and me (Oct. 7, 1993, issue), (1) has identified an inaccuracy in our response to the letters to the editor about the study (Feb. 3, 1994, issue). (2) In the fourth paragraph of our response, we reported, "Ninety-three percent of the homicides involving firearms occurred in homes where a gun was kept, according to the proxy respondents." What we should have said was that 93 percent of proxy respondents for victims of homicide involving firearms provided information about the presence or absence of a gun in the home. Sixty-two percent of this group reported that the victim lived in a home where one or more guns were kept, not 93 percent, as we stated in our reply. This figure is still substantially higher than the percentage of all case households with guns (45.4 percent) and the percentage of control households with guns (35.8 percent). (1) Although I regret the error, it does not change the validity of our original response. (2)
In our response to the letters to the editor, we predicted that our findings would be corroborated by others. (2) In 1997, Cummings and colleagues at the University of Washington reported the results of a case-control study of the association between the purchase of a handgun and homicide or suicide. (3) Their finding -- that the legal purchase of a handgun appears to be associated with a long-lasting increased risk of violent death -- is generally consistent with our results. Case-control studies of suicide in relation to gun ownership have generated similar findings. (1,4)
Arthur L. Kellermann, M.D., M.P.H.
Atlanta, GA 30322