Recent studies indicate that nearly 70% of the adult population in developed countries now use medicinal drugs, over 20% are prescribed five or more drugs, and approximately 12% are exposed to potentially serious drug-drug interactions (Fokter, Možina, & Brvar, 2010; Guthrie, Makubate, Hernandez-Santiago, & Dreischulte, 2015; Slovic, Peters, Grana, Berger, & Dieck, 2007). The day-to-day responsibility of safely using combinations of medicinal drugs often rest with lay individuals acting in non-clinical contexts without professional supervision (Britten, 2009; Friedman, Geoghegan, Sowers, Kulkarni, & Formica, 2007). Consequently, understanding that certain drug combinations present a ‘synergistic risk’ (i.e., the risk attributable to the drug combination is greater than the sum of the risk attributable to each constituent drug) is often central to avoiding harmful drug-drug interactions (Bell, 1998; Sellers, Schoedel, & Romach, 2006). We conducted two studies to test whether providing individuals with information about a drug combination that presents a synergistic (cf. additive) risk elicited higher perceived risk and, therefore, increased precautionary dosing behaviours. Participants were presented with a scenario describing how two symptoms of an infection that could each be treated by a different drug. In Experiment 1 (N = 120), information about the effects of combining the two drugs was varied: (i) no information, (ii) combination elicits an additive risk, or (iii) combination elicits a synergistic risk. In Experiment 2 (N = 445), the size of the risk (small or large) and the participant’s role (patient or doctor) was also varied. In both studies, perceived risk and negative affect increased in response to information about the increased probability of side effects from the harmful synergistic drug interaction. Despite these increases, participants did not adjust their drug dosing behaviour in either experiment: dosing was similar when these interactions were large or small, or when they had synergistic or additive effects. These findings suggest that people may struggle to transfer their knowledge of drug-drug interaction risks into decision making behaviours. Hence, both researchers and medical professionals should take care not to assume that holding accurate risk perceptions of a drug’s side effects will result in decisions that help to avoid harmful drug interactions.