There is scarce empirical evidence on the impacts of drug consumption decriminalization, especially, on problematic drug use and violence. In 2001, Portugal decriminalized the consumption of all illicit drugs. In this paper, we focus on determining the short, medium, and long-term impact of Portuguese decriminalization on mortality due to drug use and homicides, from both theoretical and empirical perspectives. We model drug consumption using an intertemporal consumption model and the decisions of trafficking firms to gain market share employing an optimization model. Our results suggest a non-linear effect of decriminalization on drug consumption risk and increasing incentives for firms to expand their market share employing violence after decriminalization. Empirically, we estimate a negative short-run effect on drug-related deaths and null long-run impacts of this legal reform. In terms of homicides, we find a positive effects in a range of 28.7%-34.2% in the medium- and long-term.