To determine risk factors among licensed firearm retailers for disproportionate sales of handguns that are later subjected to ownership tracing, generally after use in crime.
Case–control; the study period was 1998–2003. Cases were all eligible firearm retailers whose handguns were later traced at a rate that significantly (p<0.05) exceeded the expected value. Controls were a 4:1 random sample of the remainder. Data were obtained from sales and tracing records for 1998–2003 and site visits conducted August–December, 2004.
Subjects and setting
60 cases and 240 controls, from the 573 retailers in California selling ⩾ 50 handguns annually during the study period.
Main outcome measure
Status as a case. Odds ratios were used to measure relative risk.
In multivariate analyses, cases had larger sales volumes, sold inexpensive handguns more often, had a higher percentage of sales denied because the prospective purchasers were prohibited from owning firearms, and were more likely to be in an urban area, in or near a city with a policy of tracing all recovered crime guns. The effects of several risk factors, including status as a pawnbroker and sales to law enforcement personnel, appeared to be mediated by purchaser characteristics for which denied sales are a proxy measure.
A number of factors—most of them characteristics of the retailers or of their handgun purchasers, and most of them available in existing data—were linked to disproportionate sales of handguns that are later used in crime.