Getting Scientists to Tell All About Possible Conflicts of Interest

This is SCIENCE IN THE NEWS in VOA Special English.  I'm Bob Doughty. And I'm Pat Bodnar.  This week -- dealing with possible conflicts of interest in scientific research.

JAMA, the Journal of the American Medical Association, is changing its publication policy.  The editors say they are strengthening the rules as a result of recent cases involving possible conflicts of interest.

JAMA says a conflict may exist when relationships with other organizations could influence a researcher.

Those possible conflicts recently have led to corrections to a few studies that appeared in the journal.  The corrections listed ties to the drug industry that researchers had not reported.

In May, the journal published a report on the risks of rare harmful effects from two drugs used for rheumatoid arthritis.  The report said studies of the drugs, Humira and Remicade, suggested an increased risk of serious infections and cancers.

JAMA published a correction after the authors of the report provided more information about possible conflicts of interest.  The two doctors who listed a drug industry connection had more extensive ties than were reported with the study.  Still, the two doctors say they do not believe these financial relationships influenced their scientific work.

That was also what researchers involved in another study said after JAMA editors learned of their financial ties to drug companies.  That study appeared in February.  It found that pregnant women who stop taking antidepressant drugs increase their chances of becoming depressed again.  The report listed thirteen researchers involved in the study.  Later, the editors of JAMA learned that most of the thirteen had ties to drug companies that make antidepressants.

The Wall Street Journal reported that the authors failed to list more than sixty different financial links to drug companies.  The researchers told the newspaper that these links have no effect on their work.  Before the study appeared, only two of the thirteen researchers provided information about possible conflicts.

A third study that led to a correction appeared in the journal dated July nineteenth.  The study involved migraine headaches and cardiovascular disease in women.  The report did not list any possible conflicts of interest.

But a reporter informed JAMA about financial relationships between the researchers and drug companies.  The journal published a correction on its Web site on July eighteenth to list these ties.

They include research support and payments for advice and for acting as a speaker.  These are all common forms of relationships between drug companies and researchers.

The researchers say they believe they have no financial interests or relationships that are important to this study.  The journal editors disagreed.  In any case, the researchers say they have learned it is best to report all relationships with for-profit companies.  That way, they say, the publication can decide what is "relevant."

Readers might have no trouble with a study that offers good news for a company even if they know that the company paid for the study.  Or if they know that the authors of the study have close ties to the industry.

But people cannot know what to think of these relationships unless they know about them.  They need to consider all the information as they weigh the study in their mind.  If they find out later about possible conflicts, then their trust in the scientists and possibly in the publication may be damaged.  If enough situations like this happen, then science in general is the loser.

This, in short, is the argument of those who criticize researchers for not reporting possible conflicts.

And there are more immediate concerns.  The studies that appear in medical journals also help guide doctors in their treatment decisions.

For example, many doctors who treat women on antidepressants are now unsure what to advise them if they become pregnant.  The findings of the study suggested that women on antidepressants should continue to take them throughout their pregnancy.  But doctors may now wonder if they can be sure that drug company connections did not influence these findings.

This issue of trust and reporting possible conflicts of interest is not limited to publications.  Government agencies have to deal with similar concerns.  People could question the independence of government scientists who have close ties to private industry.

Drug companies have to test their products.  They can do it themselves.  Or they can seek the services of experts at universities and other research centers.  Independent testing is important especially if a drug requires government approval.

Sometimes, so many experts have close ties to companies, it can seem difficult to find some who do not.  People could argue that the situation has gotten out of control and represents a threat to public health.

But others would say business is business, even in the business of medicine.  Scientists who are investigating a new drug for a company today may have done work for a competitor in the past.

By this argument, problems exist only when researchers fail to report financial relationships that present real conflicts.

Scientists can be accused of misleading people if they do not disclose their industry ties.  Yet what might appear to be a conflict of interest to some might not to others.

As a result, it can be difficult for scientists to know what to report.  Could the gift of a medical book for speaking at training programs put on by a drug company represent a conflict?

Scientists may truly believe they have nothing to hide.  At the same time, they do not want to give people a reason to question their scientific judgment.

Studies in major publications like the Journal of the American Medical Association are peer-reviewed.  That means other researchers have to agree that a study should be published.  These reviewers have to decide that the study was designed well and that the findings have scientific value.

Since the late nineteen eighties, JAMA policy has required complete reporting of all ties related to the subject discussed in an article.  There is a special form for each member of a research team to list any possible conflicts of interest.

JAMA is amending this policy beginning next year.  Researchers will have to include all possible conflicts of interest in their article at the time they send it in for consideration.  Researchers will have to list at the end of their article all company connections or other financial support for their work.  They will be expected to include information from within the past five years and for the near future.

Journal editors say each researcher's list will be considered part of the study if the report is accepted for publication.  They say more information is always better than less.  Researchers who are not sure what to list are being told to call the journal office for guidance.

The Center for Science in the Public Interest says violators of the policy should face a three-year ban from the journal.  A JAMA spokeswoman, Jann Ingmire, says that is not likely to happen.  She says a ban could be considered illegal, a restraint of trade.

Jann Ingmire says the most important question when deciding to accept research for publication should be this: is the science good?  She says research that uses good science and study design is the one that should be published and used to guide medical decisions.

SCIENCE IN THE NEWS was written by Nancy Steinbach and produced by Brianna Blake.  Transcripts and archives of our shows are at voaspecialenglish.com.  I'm Pat Bodnar. And I'm Bob Doughty.  Our e-mail address is special@voanews.com.  We hope you can join us again next week for more news about science in Special English on the Voice of America.