How Can We Trust Clinical Studies When Scientists Have a Conflict of Interest?

Just because a scientific study exists, doesn’t make it reliable. This adage became excruciatingly clear in the December 7, 2003 Los Angeles Times exposé of officials at the National Institute of Health (NIH), who have worked as consultants for companies whose drugs were linked to the deaths of patients taking part in NIH studies. According to the report, it’s also become common for scientists who work for the NIH to consult at private companies for stock options and/or hefty consulting fees. 1

The consulting deals between drug companies and employees at the NIH, however, have gone all but unnoticed, the Times said, because the NIH allows more than 94 percent of its top-paid employees to keep their consulting income confidential.

Interviews and corporate and federal records obtained by the Los Angeles Times document hundreds of consulting payments to ranking NIH officials. Here are a few:

  • Dr. Stephen I. Katz, director of the NIH’s National Institute of Arthritis and Musculoskeletal and Skin Diseases, collected between $476,369 and $616,365 in company fees in the last decade. One company paid him more than $140,000 in consulting fees, and went on to win $1.7 million in grants from his institute.
  • Dr. Ronald N. Germain, deputy director of a major laboratory at the National Institute of Allergy and Infectious Diseases, has collected more than $1.4 million in company consulting fees in the last 11 years, plus stock options. One of the companies collaborated with his laboratory on research. The founder of another of the companies worked with Germain on a separate NIH-sponsored project.
  • Dr. Richard C. Eastman, the NIH’s top diabetes researcher in 1997, wrote to the Food and Drug Administration that year defending a product without disclosing in his letter that he was a paid consultant to the manufacturer. Eastman’s letter said the risk of liver failure from the drug was “very minimal.” Six months later, a patient, Audrey LaRue Jones, who was taking the drug in an NIH study that Eastman oversaw, suffered sudden liver failure and died. Liver experts found that the drug probably caused the liver failure.

Medical ethicists have said the consulting arrangements represent a clear conflict of interest. The problem is, a scientist who gets a lot of money in consulting fees is going to want to make sure the company paying him does very well, and sometimes that results in the death of a research participant.

Death of a patient had no impact on continuance of a study

Jamie Ann Jackson, who had been listed as “Subject No. 4” in a NIH study of the treatment of kidney inflammation related to lupus, died because of a complication involving the drug Fludara, made by Schering AG. As it turns out, Dr. Stephen I. Katz, the senior NIH official involved in the study, was a paid consultant for Schering AG. Katz could have stopped the study immediately after Jackson’s death or warned doctors outside the NIH who were prescribing the drug for similar disorders. But he did neither, because, as the Times reported, either step might have threatened the market potential for Schering AG’s drug.

According to the Times, in an article published in the May 2001 issue of the journal Pharmacotherapy, the doctors, three from Katz’s institute, wrote that Fludara “was well tolerated” and thanked the company for providing the drug and “analytical support.”

It wasn’t until December, 2003, 4 1/2 years after Jackson died, that these same doctors published a full-length article in the journal Transfusion describing her death.

“Such dual roles – federal research leader and drug company consultant – are increasingly common at the NIH, an agency once known for independent scientific inquiry on behalf of a single client: the public,” writes David Willman, author of the Times report. And, in fact, conflicts of interest among university medical researchers have also received wide attention in recent years.

Private consulting fees distort interpretation of study results

Dr. Curt D. Furberg, former head of clinical trials at the National Heart, Lung and Blood Institute, is quoted in the Times article as saying “Science should be for the sake of gaining knowledge and looking for the truth. There should be no other factors involved that can introduce bias on decision-making

“Private consulting fees tempt government scientists to pursue less-deserving research and to ‘put a spin on their interpretation’ of study results,” he adds.

How do we find credible research?

So, how can we trust the interpretations of a study knowing that scientists sometimes distort the results to favor their clients? And what about the study’s participants? How can they be sure the experimental treatments they receive are chosen on merit and not because of what a researcher stands to gain? With the amount of time and planning that goes into designing a study to stand up to scientific standards, it’s absurd to ignore the human element. The more background on the people, the institute, or organization that performed or funded the research, the better we can assess the reliability of the data.

Here are some guidelines we at Smart Publications use when assessing studies:
Who funded the study?
It’s easy to automatically disregard a study that has been funded by a company that wants to promote its products, but we realize that a company isn’t going to spend a lot of money on research that makes them look bad. The research, however, becomes more valid if the results appear in a peer-reviewed journal because that means the methods and conclusions will have been critically examined by other experts in the field. Of course, we’re skeptical if a particular scientist’s work is always favorable to the organization that sponsored the study.

  • What kind of study is it?

1) Epidemiological studies are observational studies, best at identifying powerful associations, like the one between smoking and lung cancer. They can’t establish cause and effect-such as Alzheimer’s disease is caused by cooking in aluminum pots-but rather can only suggest a relationship between two factors or events. When the relationship between factors is weaker, like the link between smoking and breast cancer, epidemiological studies are less reliable and often produce contradictory results. Also, the larger the study and the longer it has been carried out-such as the 45-year-old Framingham Heart Study and the Nurses Health Study-the more certain you can be of the findings, because researchers can better account for factors that could lead to a bogus conclusion.

2) Clinical trials use people as subjects and randomly assign people to two treatment groups. In a blind or single-blind trial the subjects don’t know whether they are receiving the treatment or a placebo until the study is completed. Placebos are treatments that appear identical to the real treatment, but contain none of the active ingredients being studied. Using placebos helps eliminate results that may be due to the subject’s expectation that something is supposed to happen. In a double-blind trial neither the subjects nor the researchers are aware of the treatment being used. This type of study provides the greatest precision because it removes any possibility of experimental bias.

3) Laboratory experiments are done in vitro on cells, or tissue samples in culture, or in vivo on laboratory animals — usually rodents. They can be tightly controlled, in that the scientists can make sure the compared groups are genetically identical and the conditions to which they are exposed (except for the factor being studied) are identical. But, as Jane Brody points out in her article, (“Personal Health: A Study Guide to Scientific Studies” New York Times, Aug. 11, 1998) “It is a long way from the test tube or laboratory mouse to man or woman. The fact that all people do not have the same genes, hormones, metabolism or life circumstances may greatly modify the effects. Or human biology may be such that whatever happens in a test tube or lower animal would not apply to us.” 2

  • Has the study been peer-reviewed?

After receiving funding and completing the research, researchers write a manuscript and submit it for publication. The journal sends out the manuscript for peer review, a process in which several scientists with expertise in the area assess its worthiness and importance. The manuscript is not accepted for publication until the reviewers are satisfied that the methods and conclusions are valid and worthwhile.


Finally, in looking at scientific studies, we realize that a single study does not mean a substance or nutrient is either beneficial or hazardous. As Jane Brody writes, “No matter how reputable the scientist or the institution where the research was done or how compelling the findings appear to be, they must be confirmed by one or more independent studies before scientists will accept them as gospel.”

Before the scientific community and public accept the benefits and advantages of a particular substance or supplement, it’s necessary for it to be studied repeatedly from various angles, by more than one research team. Science is often a guessing game, but with ethical researchers insuring that scientific measures are put in place, it is possible to get valid results that have broad and significant implications for the health and welfare of mankind.


  1. Willman, David. “Stealth Merger: Drug Companies and Government Medical Research” Los Angeles Times, December 7, 2003.
  2. Brody, Jane E. “Personal Health: A Study Guide to Scientific Studies.” New York Times, August 11, 1998.