A retrospective time trend study of firearm and non-firearm homicide in Cape Town from 1994 to 2013.

BACKGROUND: Gunshot injuries from interpersonal violence are a major cause of mortality. In South Africa (SA), the Firearms Control Act of 2000 sought to address firearm violence by removing illegally owned firearms from circulation, stricter regulation of legally owned firearms, and stricter licensing requirements. Over the last few years, varied implementation of the Act and police corruption have increased firearm availability.

OBJECTIVES: To investigate whether changes in firearm availability in SA were associated with changes in firearm homicide rates.

METHODS: This was a retrospective time trend study (1994 - 2013) using postmortem data. Time trends of firearm and non-firearm homicide rates were analysed with generalised linear models. Distinct time periods for temporal trends were assigned based on a priori assumptions regarding changes in the availability of firearms.

RESULTS: Firearm and non-firearm homicide rates adjusted for age, sex and race exhibited different temporal trends. Non-firearm homicide rates either decreased or remained stable over the entire period. Firearm homicide increased at 13% annually from 1994 through 2000, and decreased by 15% from 2003 through 2006, corresponding with changes in firearm availability in 2001, 2003, 2007 and 2011. A 21% annual increase in firearm homicide after 2010 coincided with police fast-tracking new firearm licence applications. Cape Town's coloured population experienced a significantly greater increase than other population groups following additional exposure to illegal firearms from 2007.

CONCLUSIONS: The strong association between firearm availability and homicide, and the reversal of a decreasing firearm homicide trend during a period of lax enforcement, provide further support for the association between reduced firearm homicide and stricter regulation.