Background: The expanding role of industrial support in biomedical research has resulted in both substantial interest and controversy in recent years. Our hypothesis was that, from 1985 to 2002, the role of industrial support in orthopaedic research increased, as documented by the research presented at the annual meetings of the American Academy of Orthopaedic Surgeons.

Methods: We analyzed the frequency and types of self-reported conflicts of interest for all presentations at the annual meetings of the American Academy of Orthopaedic Surgeons in 1985, 1988, 1992, 1997, 1999, and 2002. Conflicts of interest were recorded directly from the final program for each meeting analyzed. The analysis focused on the scientific presentations, Instructional Course Lectures, symposia, poster exhibits, and scientific exhibits. Information about specific types of support received by authors was first required in 1988.

Results: The incidence of conflicts of interest increased from 3% in 1985 to 39% in 2002 for scientific papers (p < 0.001); from 10% to 74%, respectively, for symposia (p < 0.001); from 22% to 60% for Instructional Course Lectures (p < 0.001); from 10% to 60% for scientific exhibits (p < 0.001); and from 9% in 1992 to 14% in 2002 for posters (p < 0.001). For presentations of all types, the incidence increased from 10% to 32% (p < 0.001). The types of conflict of interest also changed significantly from 1999 to 2002. In 1999, 73% of conflicts were documented as support directed to institutions and 27%, as support to individuals; in 2002, 57% were reported as support directed to institutions and 43%, as support to individuals (p < 0.01).

Conclusions: The role of industrial support of orthopaedic research increased significantly between 1985 and 2002, as evidenced by the increase in the self-reported conflicts of interest for all types of presentations at the annual meetings of the American Academy of Orthopaedic Surgeons. In addition, the support directed to individuals, in contrast to that directed to institutions, increased significantly.