Corporate sponsorship and research

Abstract

Can bias taint or undermine the value of medical research? A recently published article by Kace A. Ezzet, MD (J Arthroplasty 18 (suppl 3):139, 2003) examined the relationship between the source of funding for research and the reported outcomes. In his review of an extensive series of consecutive articles and scientific presentations from 2 major orthopaedic meetings and 2 orthopaedic journals, Dr. Ezzet found that industry-supported research projects had far superior positive outcomes compared to those projects that were independently funded. The obvious concern is that the funding source may introduce a bias that would limit the value of these research investigations. This strong correlation between positive findings and corporate sponsorship does not, in and of itself, imply cause and effect. For instance, it simply may be that large volume centers, by nature, attract research-oriented surgeons who, in turn, because of their leadership and reputation, are able to obtain generous commercial funding. Furthermore better outcomes are frequently more achievable at these high volume centers (Lavernia et al. J Arthroplasty 10:133, 1995), thus creating an inherent bias, but not one with a pejorative or malevolent nature. How can research-oriented surgeons cope with the phenomenon of “corporate taint”? How do we encourage the spirit of corporate giving without biasing the quality of research? First, mandatory disclosure and voluntary openness regarding funding sources allow the reader to know that a conflict of interest may exist. All peer reviewed scientific journals and open scientific meetings must insist that writers and speakers divulge any potential conflict of interest. While this is not an absolute protection, as noted by Dr. Ezzet, accountability of all parties is the first line of defense. A second important safeguard, and perhaps the most important, is the underlying integrity of the investigators. They must have a scrupulous sense of honor and a willingness to divulge poor as well as favorable clinical outcomes in a forthright manner, regardless of fiduciary cost. This important character trait allows our scientific orthopedic community to function in an efficient and productive manner. Third, the scientific design of the research project should have built-in safeguards to minimize, if not eliminate, bias. To ensure proper data design, collection, and analysis, a clinical research scientist should be intimately involved in overseeing all research studies. At the Hip Society meeting in 1999 (Malchau. Spine 25:285, 2000) Dr. Malchau suggested that the introduction of new prostheses should follow an orderly and rational pathway starting with basic science evaluation and pre-clinical trials, followed by an RSA analysis and a randomized clinical trial. Only then did he feel that general release of this new prosthesis should be allowed. While this sequence is not always possible or feasible, certainly a controlled study evaluating safety and efficacy conducted at multiple centers would be a reasonable alternative to the more common pattern seen today involving general release of a prosthesis or technique followed by retrospective reports of its ultimate efficacy. None of these strategies is perfect or absolute. Careful editorial scrutiny and critical evaluation by the readership are necessary. In the end, respecting the tenets of ethical behavior and honesty in research, regardless of the source of financial sponsorship, along with clearly delineating all conflicts of interest, would greatly enhance the well-being of our patients and the general public.