Background In Botswana, road traffic crashes (RTCs) are a leading cause of death and disability, second only to HIV/AIDS in young adult males. Alcohol has been recognised as one of the principal risk factors for RTCs, particularly for fatal RTCs, globally.

Aims/Objectives/Purpose To examine the characteristics of fatal alcohol-related crashes (FARCs) in Botswana from 2006–2010 and to examine changes in rates pre- and post-levy.

Methods This study analysed data from Botswana Police Accident Database from 2006 to 2010. Bivariate analyses examined associations between FARCs and other demographic and risk factors (eg, age, sex, injury severity, seat belt use). We calculated rates of FARCs per 10 000 registered vehicles and per 100 000 population and used z-tests to test for statistical significance.

Results/Outcome A large percentage of FARCs occurred on weekends (49%), among males (78%), and among 25–34 year-olds (35%). Pedestrians were identified as a high risk group (38%). Comparing changes pre- and post-levy, we found that there was a statistically significant change in FARCs per 10 000 registered vehicles (rate pre-levy=10.4; 95% CI 9.1 to 11.8 vs rate post-levy=8.3; 95% CI 7.3 to 9.3; p=0.01). However, rates per 100 000 population remained stable pre- and post-levy (rate pre-levy: 6.9; 95% CI 6.0 to 7.7 vs rate post-levy: 7.5; 95% CI 6.6 to 8.4; p=0.29).

Significance/Contribution to the Field This study identified important characteristics of fatal alcohol-related crashes, which can be used to tailor prevention programmes in Botswana (eg, sobriety checkpoints), and provided an assessment of the impact of the national alcohol levy.