Purpose
Generic switching of warfarin was recently repealed in Denmark, as adverse drug reaction (ADR) reports suggested risk of excessive anticoagulation following switches from branded to generic warfarin. We investigated this putative association in a formalized pharmacoepidemiological analysis.
Methods
We conducted a nationwide cohort study based on Danish healthcare registries, including data from the introduction of generic warfarin until the repeal (January 2011–April 2015). We followed Danish warfarin users over time and compared the rate of incident hospitalizations due to excessive anticoagulation (i.e. increased INR or any bleeding requiring hospitalization) in periods following a recent switch to generic warfarin to the rate in periods without a recent switch.
Results
We included 105 751 warfarin users, filling a total of 1 539 640 prescriptions for warfarin (2.5% for generic warfarin). This constituted 89.0% of all warfarin prescriptions in Denmark during the study period. We observed 19 362 switches to generic warfarin during the study period. The adjusted hazard ratio for excessive anticoagulation following a recent switch from branded to generic warfarin was 1.1 (95%CI, 0.8–1.4). The result was robust within subgroups and several sensitivity analyses.
Conclusion
Switching from branded to generic warfarin is not associated with an increased risk of hospitalization with excessive anticoagulation. However, a minor excess risk of transient INR increase cannot be excluded. Pharmacoepidemiological studies provide an effective method for swift evaluation of hypotheses generated by ADR-reports.