Should We Trust Medical Journals?

How would you feel if you found out, after your doctor had prescribed an antidepressant that he or she recommended to you, that they had financial ties to the drug company that makes the antidepressant? You probably would feel that the doctor had a conflict of interest and should have disclosed the connection while making the recommendation.

We get much of our medical information from our doctors. To keep up with advances in their fields,   the doctors in turn rely to a great extent on medical journals like the Journal of the American Medical Association (JAMA). I’ve always assumed that the studies and reviews published in journals as venerable as JAMA have been vetted thoroughly to make sure any conflicts of interest are disclosed. Unfortunately, several incidences in the past few months show that my assumption was wrong.

In February a study was published in the Journal of the American Medical Association (JAMA) which showed a high rate of depression relapse among women who stopped using antidepressants during their pregnancy.

Unfortunately, JAMA was not made aware that most of the study authors had financial ties to the pharmaceutical industry, including those companies that make antidepressants. The head author of the study, according to an article published in the Wall Street Journal last week, had “more than 60 different financial relationships with drug companies.” Of the thirteen authors, only two revealed their ties at the time the study was published. JAMA apparently didn’t spend a lot of time researching possible conflicts of interests that the study authors might have before publishing the study.

Not only did most of the study authors fail to reveal their ties to drug companies, but several, subsequent to the publication of the study, went on the lecture circuit and attacked studies that have suggested that antidepressants can be harmful to the developing fetus.

Last week a problem with a favorable review of the Vagus Nerve Stimulation (VNS) device, used to treat depression, was made public. The review was in an influential medical journal, Neuropsychopharmacology , whose editor is Dr. Charles Nemeroff. What was the problem? It’s kind of like that commercial, “I’m not just the Hair Club for Men president – I’m also a client.” Dr. Nemeroff not only is the editor of the journal, he’s one of the authors of the favorable review and – surprise – he has financial ties to the company that makes the VNS device, Cyberonics.

It’s important to be skeptical of every study that you read, and take the findings with a grain of salt. It’s very important to know whether the study is funded by a company or organization that has something to gain other than the desire to gain knowledge.

For me, while this disclosure about the pregnancy and depression study doesn’t completely invalidate its findings, it means that I will take them not with a grain of salt, but a huge honking 50 pound bag of it. I never had a whole lot of faith in the effectiveness of VNS after reading several articles that tore the lone study supporting it apart, like this one on the Public Citizen website, so the exposure of the conflict of interest doesn’t change my view of its effectiveness.

In a perfect world, a study author having ties to drug company or companies would not slant a study’s results in one direction or another. But let’s get real. These studies are not done by robots – they’re done by human beings, who can consciously or subconsciously be biased in a certain direction. Bear in mind that I am being charitable in assuming that these researchers did not deliberately slant the study results to benefit the drug companies they had relationships with.

In its story about America Online last week, Time magazine disclosed that they are both owned by Time Warner. This is standard practice in mainstream journalism, whether the piece is flattering to the parent company or not. Knowledge of the affiliation allows us to evaluate the piece’s objectivity in an informed manner. Shouldn’t any information written about our health also be required to automatically disclose all and any associations that might have biased the writer? It seems obvious to me. Hopefully, medical journals will adopt more stringent standards and take it upon themselves to do some research into the backgrounds of study participants. Then we and our physicians should be able to trust them again.