Scientific journals and their authors' financial interests: a pilot study

Abstract

Background: The credibility of modern science is grounded on the perception of the objectivity of its scientists, but that credibility can be undermined by financial conflicts of interest. The US Public Health Service and the National Science Foundation issued regulations effective October 1, 1995, regarding the disclosure of financial interests in the submission of grant proposals. Several scientific journals have also established pertinent policies for authors and editors. The objectives of this study were: (1) to select a set of published articles and observe the degree to which a sample of authors hold a financial interest in areas related to their research that are reportable under current standards, and (2) to examine the hypothesis that significant numbers of authors of articles in life science and biomedical journals have verifiable financial interests that might be important for journal editors and readers to know. This paper measures the frequency of selected financial interests held among lead authors of certain types of scientific publications and assesses disclosure practices of authors and journals.

Method: These objectives were applied to a pilot study of Massachusetts academic scientists who were cited as first or last author in at least one article published in 1992 in 14 leading journals of cell or molecular biology and medicine. We created a database of every original article published in 1992 by 14 leading life science and biomedical journals, supplemented by data sets consisting of (1) Massachusetts biotechnology firms, including their officers and scientific advisory boards, and (2) scientists listed as inventors on patents or patent applications registered with the World Intellectual Property Organization.

Results: We examined 1,105 university authors (first and last cited) from Massachusetts institutions whose 789 articles, published in 1992, appeared in 14 scientific and medical journals. Authors are said to 'possess a financial interest' if they are listed as inventors in a patent or patent application closely related to their published work; serve on a scientific advisory board of a biotechnology company; or are officers, directors, or major shareholders (beneficial owner of 10% or more of stock issued) in a firm that has commercial interests related to their research. Applying the criteria to the reference population of journals and Massachusetts academic authors, we measured the following frequencies for lead authors: 0.20 for serving on a scientific advisory board; 0.07 for being an officer, director, or major shareholder in a biotechnology firm, and 0.22 for being listed as an inventor in a related patent or patent application. The joint frequency of articles in the journals reviewed with a lead author that meets one of the three conditions is 0.34.

Conclusions: One of every three articles in our sample has at least one Massachusetts-based author with a financial interest, and 15% of the authors in our sample have a financial interest relevant to one of their publications. For the year 1992, the rate of published voluntary disclosures of financial interest (as defined in our study) is virtually zero, but relatively few scientific and biomedical journals at that time required any such disclosure to journal editors and reviewers. Further research is needed to determine the effectiveness of mandatory disclosure requirements by some journals.