In December 2004, the Journal of the American Medical Association(JAMA) published a meta-study examining the frequency reported outcomes of clinical trials that differed from study protocols and data (Chan et. al Empirical evidence for selective reporting of outcomes in randomized trials: comparison of protocols to published articles.
JAMA. 2004;291(20):2457-2465. doi:10.1001/jama.291.20.2457.) The authors reported:
"[Of] One hundred two trials with 122 published journal articles and 3736 outcomes were identified. Overall, 50% of efficacy and 65% of harm outcomes per trial were incompletely reported."
"...In comparing published articles with protocols, 62% of trials had at least 1 primary outcome that was changed, introduced, or omitted. [86%] percent of survey responders (42/49) denied the existence of unreported outcomes despite clear evidence to the contrary."
Today, a Wall Street Journal front page story, "Worrisome Ailment in Medicine Misleading Journal Articles" (May 10, 2005) focuses on the same problem. The article presents the responses of various journal editors to questions about the problem of journalistic integrity, before concluding that journals are all addressing the problem and striving to present more balanced studies. Do the Wall Street Journal's assurances settle the issue?
Briefly, publishing has multiple points of responsibility. Scientists in the lab from techs to post-docs and residents report up to the principal investigator, who oversees and sets the direction of the lab research and assumes responsibility as an author on the paper. Publishing authors should be forthright about their results, and journal editors and publishers must select articles that compel readers and sponsors to support the magazine, while at the same time maintaining journalistic integrity. The multiple goals of all the players aren't always easy to balance.
The issue at stake is that the medical journals are increasingly dependent on advertising from pharmaceutical companies, and are therefore increasingly beholden to their influence. Journal editors would trust sponsored drug efficacy studies but for the reality that a pharmaceutical company has every incentive to report positive (and therefore sometimes biased) results. Journals in effect receive advertising support from the very same authors they are publishing.
What is most clear from the WSJ article is that at every juncture all parties both recognize the problem and at the same time shrug off their own responsibility in the solution. There are two problem areas, according to the WSJ account:
1) The editorial challenge of verifying submitted research.
2) The tension between the pharmas and journals about which results and conclusions are valid.
Pharmaceutical companies feel they are being unfairly scrutinized. All parties disagree about who has greater responsibility. WSJ quotes JAMA's editor in chief, who says: "The single thing we change the most often is the conclusion....It comes in as: 'This product is the greatest thing' and we say, 'Under these circumstances, in this population, this medications seems to control a,b and c.'"
JAMA's deputy editor adds: "Science depends on trust...[b]ut if you have trust, you're going to be fooled. You can't have a policeman in every lab." He suggests that the problems stem from dishonest research, which is out of the hands of editors.
The editor of Arthritis and Rheumatism quoted by WSJ is willing to let others to sort out the fray:
"People have to be realistic about what a journal can do." He said that the job of verifying the drug data was the job of the FDA.
An associate vice president of PHARMA (The Pharmaceutical Research and Manufacturers of America) complained: "Why are our submitted articles different from all other submitted articles?"
But when the FDA criticized article authored by Merck-Johnson & Johnson on the pros of selling a drug over the counter, a Merck official was very critical: "[The FDA] went very, very strictly by the label."
Depending on who you talk to, it's either the actual researchers doing the experiments that are at fault, the interpretation of the data by the authors, the journal editors who aren't scrutinizing papers closely enough, or the FDA.
The Wall Street Journal article is interesting, but focuses undue attention on the "tension" between the pharmas, pharma researchers and the journals and therefore fails to uncover an important part of the story. Undeniably, university researchers AND journal publishers in the past 50 years are more reliant on pharma funding and partnerships. Consequently the integrity of study results (and not just those from pharmaceutical companies) should be brought into question, shouldundergo greater scrutiny.
Journal editors may indeed be increasingly challenged to adequately check the veracity of the research studies that are submitted. Without trivializing these issues however, there is another side to the story, which is not the tension between the researchers or journals and pharmaceuticals, but the symbiotic and increasingly intertwined relationship between the journals and the pharmaceutical companies. The story beyond the on-cue responses quoted would focus on the story that the WSJ chose to omit. It is not a new story.
In a 2003 journal article by Alexander Tsai in the International Journal of Health Services titled "Conflicts Between Commercial and Scientific Interests in Pharmaceutical Advertising for Medical Journals", "A Doctor's World; Inside Medical Journals, A Rising Quest for Profits"): "drug companies stopped advertising, costing the journal $1 million to $1.5 million, Dr. [Susanne] Fletcher [editor of the Annals at the time] said".
The ensuing turmoil at the journal highlighted the increasingly co-dependent relationship between medical journals and pharmas and underscored the diverging goals of editors and publishers. The NYT article elaborated on the rising conflicts of interest between successful - meaning profitable - medical publishing, where the pharmaceuticals were the source of most advertising dollars, and content integrity. The change was noticeable:
"These societies were kind of frumpy and esoteric, and they are no longer that way," Dr. Fletcher said. "They are big businesses, and editors who traditionally worked with small staffs and had time to read a lot now go to meetings about new ventures."
While the mission of medical journals to disseminate current, useful information to doctors may not have changed in essence, the advertising revenues grew astronomically. According to the New York Times report:
"...the journals have increasingly become cash cows for the medical societies and companies that own them, with annual profits in the tens of millions of dollars, largely from drug company advertisements....The New England Journal of Medicine [reported it's earnings in] 1979, when its earnings were less than $400,000 a year. Now [it's] earnings are estimated to be at least $20 million a year..."
The Times went on to quote Dr. Suzanne W. Fletcher, who until 1994 was the co-editor of The Annals of Internal Medicine, owned by the American College of Physicians (she is currently a Harvard professor):
"The journals' success has created an enormous new market force that is beginning to dominate the way medical societies are run," Dr. Fletcher said. In some not-for-profit societies, the business staff is gaining influence over editors, creating frictions between the two traditionally separate sides..."
So the question is, if medical journals are dependent on pharmaceutical companies for advertising dollars, can they afford not to be biased? For a journal, it seems that it becomes not merely a question of publishing the latest research, but a question of what the journal publishers and editors need to do to balance obligations to readers, doctors, and advertising sponsors -- to publish the latest findings responsibly while meeting the revenue expectations.