Daniel Sabanés Bové, Heidi Seibold, Anne‐Laure Boulesteix, Juliane Manitz, Alessandro Gasparini, Burak K. Guünhan, Oliver Boix, Armin Schüler, Sven Fillinger, Sven Nahnsen, Anna E. Jacob, Thomas Jaki
Open-source statistical software development is increasingly the preferred solution for leveraging new statistical methods in the pharmaceutical industry. However, with a long history of relying on licensed analysis software, there are philosophical and organizational barriers to overcome. In particular, the sustainability, reliability, usability, and feasibility of maintaining open-source statistical software long-term must be ensured. Here, we describe the open-source revolution that is emerging in the pharmaceutical industry and how it facilitates greater scaling of innovative analytical methods in statistics. We discuss challenges to open-source software adoption and propose mitigation strategies. Furthermore, we illustrate the potential for open-source software development with examples of successful projects, which highlight the roles of cross-company collaboration, career paths, education, and community building.