|
Abstract:
|
The last 50 years (and especially the last 10-15) have seen a burgeoning in collaborations among academic, industry, and government statisticians. Industry statisticians have faced increasing pressure to develop and utilize more efficient statistical techniques while controlling costs, while academics have faced increasing pressure to obtain external research funding support. Fortunately, many problems of mutual research interest have created a "win-win" in this environment, with academics supervising industry-supported PhD students efficiently developing statistical methods and software as part of their dissertations. In this talk I review several examples of successful collaborations from my own work in the design and analysis of Bayesian adaptive clinical trials, including applications in medical device trial design, adaptive borrowing of strength from historical data, network meta-analysis to assess a drug's competitive position, novel methods for rare and pediatric diseases, and Bayesian approaches for subgroup identification. While agreements over publication and intellectual property rights remain a challenge, the model has worked well and even has potential for expansion.
|
