Treatment for oral candidiasis in warfarin users may be complicated by drug-drug interactions (DDIs) between warfarin and topically applied antimycotics. However, current knowledge of these putative DDIs is merely based on case series. We therefore performed a cohort cross-over study with the objective to evaluate the potential DDIs between warfarin and miconazole oral gel or nystatin oral solution. The cohort consisted of individuals using warfarin in the period of 1998-2012 (n ≈ 7400). We collected data on cohort members' measurements of the international normalized ratio (INR) from a clinical database, and obtained information on their use of topically applied miconazole and nystatin from a regional prescription register. Potential DDIs were assessed by comparing INR values before and after initiation of an antimycotic drug. Among 17 warfarin users exposed to miconazole oral gel, the mean INR increased from 2.5 (95% CI: 2.1-2.8) to 3.8 (95% CI: 2.8-4.8) after exposure, corresponding to a mean INR increase of 1.4 (95% CI: 0.3-2.4). Among 30 warfarin users exposed to nystatin oral solution, the mean INR was 2.7 (95% CI: 2.3-3.1) before and 2.5 (95% CI: 2.2-2.9) after exposure. In conclusion, we found evidence supporting a clinically relevant drug-drug interaction between warfarin and miconazole oral gel. In contrast, we did not find any indication of an interaction between warfarin and nystatin oral solution. Nystatin rather than miconazole should be preferred when treating warfarin users for oral candidiasis.