This study used national data and a matched case-control design to estimate the relative risk of death by an unintentional gunshot associated with having firearms in the home. A sample of adults who died in the United States in 1993 from unintentional gunshot injuries was drawn from the National Mortality Followback Survey (NMFS) (n=84). Twenty controls were sought for each case from the 1994 National Health Interview Survey (NHIS) and matched to the cases by sex, age group, race, and region of residence (n=1451). Subjects were classified as having or not having guns in the home based interview responses. The relative risk of death by an unintentional gunshot injury, comparing subjects living in homes with and without guns, was 3.7 (95% confidence interval (CI)=1.9-7.2). Adjustment for covariates resulted in little change in the effect estimates. There was evidence of a dose-response effect: compared to subjects living in homes with no guns, the relative risk was 3.4 (95% CI=1.5-7.6) among subjects with one gun and 3.9 (95% CI=2.0-7.8) among subjects with multiple guns in the home. Having handguns in the home was associated with the largest effect estimates. Tests of homogeneity showed that the effect estimates did not vary significantly across categories of the matching variables. Firearms in the home appear to be a risk factor for unintentional gunshot fatality among adults. The magnitude of the observed effect estimates should be compared with those from additional studies.