Abstract
OBJECTIVES. Many policies seeking to limit handgun violence rest on the largely untested "crime gun hypothesis," which holds that subclasses of handguns differ in their risk for use in violent crime. This study tests that hypothesis for handguns used in homicides of law enforcement officers and describes the population of homicide-involved handguns. METHODS. A cross-sectional study was done of civilian (criminal) handguns used in homicides of law enforcement officers from 1980 to 1989. Life tables were generated for each year's cohort of new handguns to estimate gun-years at risk, analogous to person-years, for rate and relative risk calculations. RESULTS. Four hundred thirty-five deaths involved 428 civilian handguns. Revolvers were at greater risk than pistols. For both, risk was lowest for .22-caliber handguns. Risk was greatest for .32-caliber pistols and .38-caliber revolvers. Forty-six percent of handguns had a barrel length of 3 in or less. CONCLUSIONS. Subclasses of handguns differ substantially in their risk for use in fatal shootings of law enforcement officers. Such epidemiological data may be useful in formulating efforts to prevent these and similar instances of firearm violence.

This publication has 9 references indexed in Scilit: