Open Source and Commercialized Open Source: Risks, Mitigations, and Upsides for Commercial Regulated R&D

• Activity Number: 100
• Type: Invited
• Date/Time: Monday, August 5, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Section for Statistical Programmers and Analysts
• Author(s): Anthony Joseph Rossini and Vincent Carey*+
• Companies: Novartis Pharma AG and Harvard Medical School
• Keywords: Open Source ; statistical practice ; pharmaceutical R&D ; computer systems validation ; software ; data analysis

Abstract

Software, whether open source or commercial, needs to be assessed for the risk that it introduces to business processes. "open source" describes a license model, not a risk mitigation model. With the increasingly recognized (by non-statisticians) of the importance of statistical input and thinking, this input introduces, in addition to tremendous benefits, risks surrounding the "process of statistical analysis", which includes statistical software selection. We discuss issues surrounding statistical software selection, impact on the form and quality of the "process of statistical analysis". This is a critical but often downgraded aspect of "Statistical Practice", and we hope to provide clarity on this issue and justify why it requires careful consideration as the "process of statistical analysis" evolves from an early young (and being evaluated) to a mature stage and an essential (and streamlined) business process.