Essays on the Drug Discovery Innovation System

摘要

Despite the increase in scientific, monetary, and human resources devoted to drug discovery R&D, the annual number of FDA-approved drugs has been almost unchanged for decades. This study seeks clues to the R&D productivity paradox by exploring the drivers and barriers to innovation in the pharmaceutical industry. The main focus is on the contribution of science and knowledge form external sources to innovation.;The first essay uses patent data and provides evidence in support of the "foundational view" of creativity that a deep grasp of domain knowledge is most important in developing new ideas.;Patent-based models of firm inventive output indicate a curvilinear (first negative then positive) pattern for the contribution of codified scientific knowledge to inventive output. This finding implies that firms need to assimilate and utilize scientific knowledge across multiple inventions. The use of diverse knowledge components in innovation is positively related to quality-weighted inventive output. However, a firm's reliance on knowledge components different form own inventions, negatively influences both inventive output and quality-weighted inventive output.;Essay 2 examines breakthrough innovations (i.e. Orange Book patents) for clues about what makes inventions underpinning a drug product different from other inventions. The results indicate that inventions with fewer applications (i.e. less general or with citations across different technology classes) have a higher probability of entering the Orange Book. Broader legal scope also boosts the probability of an invention being an Orange Book patent. This finding might be caused by inventors carving out larger legal protection for more valuable inventions. In terms of knowledge recombination, higher technological knowledge diversity reduces the probability of an invention being an Orange Book patent, while technological knowledge distance (i.e. not coming from the patent's technology class) increases the probability of being an Orange Book patent. These results are in line with the "tension view" of creativity, which emphasizes the need for distant or diverse knowledge for generating new ideas.;The third essay draws on a survey of experts that explores the drivers and barriers to innovation from their perspective. The respondents possess nuanced knowledge of broad R&D spending and drug approval trends. Some barriers to innovation that they perceive are accentuated by the "molecular reductionist" drug discovery paradigm. Moreover, the essay provides evidence of several systemic failures. Lack of change in the fundamental rules of the game has created a "lock-in/path dependency failure" in which the innovation system has failed to adapt expeditiously. Deficiencies in firm capability development have lead to "transition failures". Hard (i.e. regulatory) and soft (i.e. cultural) institutional failures, along with "regulatory capture" are also evident in the responses and other data. In line with the "foundational" view of creativity, respondents ranked "depth of knowledge" higher than "diversity of knowledge" for innovation. The latter result is in line with the insights form essay 1.;These systemic failures provide a starting point for formulating policy interventions to improve the innovative output of the drug discovery innovation system.