The pharmaceutical industry has become a major source of funding for biomedical research. Investigators are now required by many journals to declare in writing any potential conflict of interest that may threaten their impartiality. While we fully support the need to disclose relationship details that would not otherwise be obvious to the readers, we are also interested in author-sponsor relationships that would constitute direct and irrefutable conflicts of interest. It has been our general observation that sponsoring company employees have been appearing as co-authors of these publications with increasing frequency.
The primary objective of this study was to characterize author-sponsor affiliation trends over time. Secondary objectives include assessments of research funding overall and with respect to time, author affiliations per journal over time and clinical outcomes per funding type over time.
Author-industry affiliations were analyzed for 500 randomly selected clinical trial publications appearing in 5 influential medical journals over a 20-year period. For each journal, 10 randomly selected issues in a given year were scanned to identify papers meeting the publication inclusion criteria. For those papers meeting these criteria, 5 articles per journal-year were randomly selected and reviewed. Information regarding author affiliations, study drug funding sources, study outcomes and other factors was recorded.
Industry provided funding, solely or in part, for 36% of the 500 clinical trials analyzed. Peer-review funding was identified in an additional 36% of the studies, whilst the remainder had no funding source declared. The number of industry-sponsored trials increased over the 20-year period of the study to a maximum of 69% of all trials reviewed for the final 4-year period reviewed (1997-2000). The number of peer-review funded trials remained constant (~20% of total studies reviewed), while the incidence of studies with no funding declaration declined with time. Author affiliation with industry increased from under 10% to a high of 66% in the final 4-year period for those trials with industry as the sole funding source. Over the 20-year study period, an increase in the incidence of studies with author affiliation with an industry sponsor over time was observed for all five journals reviewed. This increase was greatest with the Annals of Internal Medicine, Journal of the American Medical Association, and New England Journal of Medicine. Regardless of funding source, the majority of clinical trials published in the five journals reviewed demonstrated clinical outcome results that favoured the study drug. Over the 20-year period, the incidence of trials reporting positive outcomes for the study drug modestly increased for both industry sponsorship groups, while the reverse was observed for the peer-review and unfunded groups.
To our knowledge, this is the first published report to characterize author-sponsor affiliations according to funding source and time. Our results indicate that pharmaceutical industry sponsored trials are common and have increased in incidence over the past two decades. Our study indicates that author-sponsor affiliations are also common, and have increased during this same time period. These increases in industry involvement increase the potential for bias in clinical trial publications.