A method for understanding generic procurement of HIV medicines by developing countries with patent protection

Patent protection on medicines may frustrate access by blocking generic competition. Nevertheless, circumstances may still allow for generic procurement to occur anyway, especially for humanitarian cause. But to what extent does this occur? And which legal flexibilities may facilitate such procurement? We attempted to design a replicable methodology that involved linking antiretroviral (ARV) patent data (1260 patents for 12 medicines) from a World Intellectual Property Organization patent study on the 2013 World Health Organization's (WHO) Model List of Essential Medicines to all available matching procurement records in the WHO's Global Price Reporting Mechanism. We then cross-referenced these with lists of legal flexibilities which facilitate generic access where patents have been granted (e.g., supplier companies' patent non-enforcement policies, voluntary and compulsory licenses) to estimate plausible relevance. The patent data corresponded to 1924 generic procurement transactions (1.34 billion units) from 85 countries. While patents were relatively less common in these countries (the median coverage was 20%), over half (53%) of the generic procurements nevertheless aligned with patent protection in the exporting and/or importing country. The disproportionately high relevance of patents despite their lower numbers can be explained by their presence in key medicine-exporting countries and/or those with larger populations. We noted, however, that developing countries still seemed able to buy generic versions of these essential ARVs. A combination of legal flexibilities may have played important roles, but voluntary licensing agreements (VLs) between originator companies and generic ones appeared to align with the largest volumes of generic procurement where we estimated patent protection. If true, VLs may warrant proportionate attention from observers as a heavily relied upon international mechanism for facilitating generic access so that the implications can be better understood; however, we hope others repeat similar studies to investigate whether these results hold with different methodologies and samples of patented medicines, contexts, and timeframes.