Background Road traffic deaths are a substantial barrier to population health improvement in low-income and middle-income countries (LMICs). In South Africa, the road-traffic injury mortality (RTM) rate of 27 per 100 000 population is twice the global average, over 60% of which are alcohol-related. Recent US studies suggest the Uber ride-sharing service may reduce alcohol-related RTM, however RTM burden in the USA is relatively low and transport behaviours differ from LMICs.

Methods Using certification data from all deaths occurring in South Africa in the years 2010–2014 (n=2 498 216), we investigated the relative change in weekly road traffic-related death counts between provinces which received Uber services (beginning in 2013) against those that did not using a difference-in-differences approach.

Results Weekly road traffic-related deaths in provinces with Uber were lower following Uber introduction than in comparison provinces without Uber. The effect size was larger in the province which had Uber the longest (Gauteng) and among young adult males (aged 17–39 years). However, the absolute effects were very small (<2 deaths per year) and may coincide with seasonal variation.

Conclusions Overall, findings did not support either an increase or large decrease in province-level road traffic-related deaths associated with Uber introduction to South Africa. More localised investigations in South Africa and other LMICs are needed.