August 2000 Bulletin

Are researchers and industry too cozy?

Financial interests may influence journal evaluations of research

By Cynthia Shewan

The research endeavor is often couched in terms like "the search for the truth" or the "unbiased presentation of results." To answer questions, investigators undertake research studies, based on sound research designs and appropriate data analyses, to produce unbiased results in which readers can have confidence.

On the surface, this seems to be a pretty straightforward process. However, it is not always that simple. People expect they can bank on what the researcher reports. Can they?

What is the issue?

Researchers can have interests, which may influence or may be perceived to influence their reporting. Referred to as "conflict of interest," this is a condition, not a behavior. It is common.

Having a financial interest in the company whose product you are evaluating or receiving funding from the company whose product you are evaluating are the most frequently cited examples of conflict of interest.

There are also other financial arrangements that may constitute a conflict of interest: acceptance of gifts, serving as a consultant, serving on an advisory board, royalty agreements, etc.

How did this come about?

In the past, clinical researchers and industry were separate, but the two have become increasingly intertwined over the recent years. The need for research funding and the Technology Transfer Act are among some of the reasons. However financially beneficial such liaisons, Marcia Angell, MD, as former Acting Editor-in-Chief of the New England Journal of Medicine (NEJM), indicates that such cooperation should be at arm’s length, with both sides maintaining their own standards and ethical norms.

Why has breaching the separation of academic medicine and for-profit industry happened? Two reasons cited are: to facilitate technology transfer and to provide money.

At the same time that academic medicine is increasingly allying with industry, industry is seeking partnerships elsewhere, such as with contract-research organizations (CROs), which can deliver results of clinical trials faster and more cheaply than academic medical centers.

Who does it affect?

The answer is industry, individual physicians/scientists, academic institutions/medical centers and CROs.

Some universities are more stringent than others regarding faculty’s opportunities for financial ties with industry. Harvard University has one of the most stringent conflict of interest policies and, after a recent review, decided to retain its policy as is rather than become more lenient. Dr. Angell advocates that more institutions should adopt more stringent policies so that the standard is equally high rather than low and spiraling toward the bottom. As Dr. Angell so aptly put it: "House officers should buy their own pizza, and hospitals should pay them enough to do so."

Is conflict of interest a problem?

Whether conflict of interest is important and needs to be seriously considered is a source of many debates. One group would answer the question, "No." Some would say that having a financial interest in or receiving funding from the company whose product you are evaluating does not matter. The scientific process was followed and the results are the results.

Others would say that conflict of interest may be a problem. Solutions offered are: disclosure or prohibition. Prohibition does not seem feasible in this day and age.

Some support the viewpoint that having researchers identify their conflicts of interest and fully disclose them takes care of the issue. Informing readers of the conflict is sufficient.

Do authors disclose their conflicts or potential conflicts? Do journals require and enforce authors to do so?

A 1999 article in British Journal of Medicine published data revealing that many authors in their journal publications (68 of 70) had not disclosed their financial relationships with the companies whose products they were evaluating. Apparently, the journals either did not require disclosure or did not enforce it.

Mandatory disclosure might help; however, let’s look at what truth in disclosure tells us. Many authors sign that they do not have a conflict of interest when they do. Less emotional language such as "competing interests" may help with compliance.

More stringent disclosure through broadening disclosure to areas beyond the financial, such as personal, political, academic, and religious, is another possibility.

Consistently enforcing the disclosure policy by journals is a third possibility. NEJM recently found that in 18 instances it did not comply with its policy, by accepting articles for publication in which the author had a major financial connection with the company whose product was being evaluated.

Evidence is accumulating on the influence of conflict of interest:

There is mounting evidence that conflict of interest has an impact on the conclusions reached by papers in medical journals. This influence can take place at several points in the research process. Thomas Bodenheimer (NEJM, May 18, 2000) reported that industry influence can bias the research in the trial design, data analysis, publication or non-publication of the results, control over publication, and authorship (nonwriting author — nonauthor writer syndrome).

How is the issue being addressed?

The recent death of a patient in a gene therapy trial, which was found to have lapses in its protocol for patient safety, has prompted the Clinton administration to propose a broad plan to address the issue. The plan proposes to:

This plan would affect both the National Institutes of Health (NIH) and the Food and Drug Administration (FDA) in their regulations related to human research. The NIH and FDA currently have guidelines related to patient safety and conflict of interest, which can be found on their web sites.

Another solution offered is to increase the independence of investigators to conduct and publish their research.

Cynthia Shewan, PhD, is director of the AAOS Department of Research and Scientific Affairs


Angell, M. Is academic medicine for sale? N Engl J Med 2000; 342 (20): 1516-1518.

Angell, M., Utiger, RD., & Wood, AJJ. Disclosure of authors’ conflicts of interest: A follow-up. N Engl J Med 2000; 342(8).

Bodenheimer, T. Uneasy Alliance: Clinical investigators and the pharmaceutical industry. N Engl J Med 2000; 342 (20): 1539-1544.

O’Harrow, Jr., R. Harvard won’t ease funding restrictions. The Washington Post, Friday, May 26, 2000, p. A16.

Smith, R. Beyond conflict of interest: Transparency is the key. Br Med J (International) 1999; 31: 291-292.

Stolberg, SG. Fines proposed for violations of human research rules. The New York Times, Wednesday, May 24, 2000.

Winslow, R. Drug-study accounts are called overly positive. The Wall Street Journal, Thursday, June 1, 2000, p. B2.

Home Previous Page