Response to 'Conflicts of interest and independent data analysis in industry-funded studies

Abstract

To the Editor: In their Editorial, Dr Fontanarosa and colleagues1 discussed JAMA policies regarding conflicts of interest and the role of sponsors in funded studies. They addressed requirements for data analysis, which are reflected in the current Instructions for Authors: "For industry-sponsored studies ... an independent data analysis must be conducted by statisticians at an academic institution with access to the raw data set, rather than only by statisticians employed by the company sponsoring the research."2

We are writing as representatives of the International Biometric Society, which comprises approximately 5000 members devoted to development and application of statistical methodology in the life sciences. We are troubled by this policy’s asymmetric treatment of potential conflicts of interest, with data analysis that is submitted by industry requiring independent verification or replication, but no such requirement for studies submitted by other researchers. Conflict of interest is possible in all settings, and the review process should ...