Compulsory licensing (CL), provided by the Agreement on Trade-Related Aspects of Intellectual Property Rights, enables countries, including high-income countries, to ensure the protection of public health in the context of stringent intellectual property regimes. This study investigated associations between the time to attempted CL and a series of factors at the drug and country level. We used a dataset of all attempted CL that occurred from 1995 to 2014, calculated the duration as the difference in years between the year of global marketing of a certain drug and the year of attempted CL in a certain country, and applied a nonparametric event history model. We found that the Doha Declaration was quite effective in shortening the time to attempted CL. Additionally, even though global justice movements have encouraged some countries to attempt CL for various medicines since 2012, the time to CL attempts in this period became significantly longer compared to those that occurred immediately after the Doha Declaration. Our findings show that the subject of CL has not successfully expanded to oncology medicines from HIV/AIDS medicines and that recently approved medicines are not yet subjects of CL. Furthermore, our duration model suggests a learning-by-doing effect in attempting CL: previous experience of CL not only triggers CL for the same drugs in other countries but also accelerates CL for other drugs within the country.
