Deception in Mainstream Medical Journals
"The public will choose to believe a simple lie in preference to a complicated truth."
Alexis de Tocqueville
"It is better to know nothing than to know what ain't so."
Josh Billings, 19th Century humorist
In 2005 the British Medical Journal (BMJ) reported a study of 519 clinical trials in 553 publications, with 10,557 outcomes identified. Only 69% of the surveys responded with information on unreported outcomes, and these were often unreliable. For 32% of those who denied the existence of such outcomes there was evidence to the contrary in their publications.
Conclusion: "Incomplete reporting of outcomes within published articles of randomised trials is common and is associated with statistical non-significance. The medical literature therefore represents a selective and biased subset of study outcomes..." In other words, a third or more of the authors were lying; many more were engaging in deception.
Another appalling example of misleading medical reporting, ghostwriting of medical articles in prestigious medical journals, has once again been unmasked, this time, by the journals themselves. What would ghostwriting have to do with incomplete or inaccurate reporting of outcomes? Answer: Plenty.
According to a study released by the Journal of the editors of The Journal of the American Medical Association (JAMA) at an international meeting of journal editors in Vancouver and reported in the New York Times, in five of six of the top medical journals*, in over 630 articles last year, an average of 5%-11% of the authors acknowledged contributions by people whose work should have qualified them to be named as authors on the papers but who were not listed. "In the scientific literature, ghostwriting usually refers to medical writers, often sponsored by a drug or medical device company, who make major research or writing contributions to articles published under the names of academic authors."
The kicker was the nature of the JAMA study, an online questionnaire which merely solicited responses from authors. This type of study has terrific potential for reporting bias because respondents were not chosen randomly, and the authors could elect not to answer (all) the questions. Since the authors were asked to disclose their own behavior, this was an obvious opportunity to underreport the use of a ghostwriter, which is considered an academic crime akin to plagiarism. The average figure of 5%-11% of "outside contributions" to medical articles in these journals is exceedingly suspect because some of the same researchers thirteen years ago also sent out questionnaires to authors of articles published in 1996 in three of the same publications. Even though the study methods were not strictly comparable to the new study, the results were even more impressive: (admitted) ghost authorship rates were 16% in the New England Journal, 15% in the Annals, and 7% in the JAMA.
These studies in their limited methodologies have to be assumed to represent only the tip of the iceberg. Sending out questionnaires to any author of a medical paper asking if he had possibly committed fraud in the form of undisclosed help is a disingenuous tactic in the first place. It's surprising the investigators got so many honest replies. The concern, the researchers said, is that "the work of industry-sponsored writers has the potential ... to affect treatment decisions by doctors and, ultimately, patient care." Really?
Ginny Barbour, chief editor of PLoS Medicine, the journal of the Public Library of Science, said after the meeting. "We are a journal that has very tough policies, very explicit policies on ghostwriting and contributorship, and I feel that we’ve basically been lied to by authors." It's obvious that Congressman Joe Wilson had the wrong target when he attacked the President.
What to do about this problem of journalistic prevarication and fraud? Stay tuned.
*The JAMA, the New England Journal of Medicine, the Lancet, the Annals of Internal Medicine, PloS Medicine.
Martin F. Sturman, MD, FACP
Copyright 2009, Mathemedics, Inc.
EasyDiagnosis is an automated online service that analyses existing medical symptoms and predicts likely causes and conditions. Click here to find out more about this unique service and click here to try one of our modules for free.