More Journal Paper Retractions: A Sign of the Times?
Retractions happen. Sometimes the authors themselves find honest errors; at other times, retractions come after a lot of debate and a failure to replicate the data; and more and more (it seems) it’s, um, worse. Take this headline from the blog Retraction Watch: “Author retracts weight loss surgery paper after admitting most, if not all, of the subjects were made up.” That’s the type of headline that keeps editors awake at night. Is there really a problem? If so, is there a fix? Or, at the very least, is there any preventive therapy?
For the cardiology community, the issue has been forced into awareness with ugly cases in recent months and years, including top-tier journals. The sense is that doctors are infecting their research papers with invented data, bogus patient surveys, negligent record-keeping, manipulated images, and lack of informed consent—leading to retractions and, in some cases, outright firings.
Carl Zimmer’s The New York Times article1 noted that the editor-in-chief of Infection and Immunity discovered paper “doctoring” by a Japanese researcher that ultimately led to a swarm of retractions: half a dozen articles were pulled by Ferric C. Fang, MD, from his own journal and at least 30* others from other journals.
Dr. Fang, professor of laboratory medicine and microbiology at the University of Washington School of Medicine, and fellow editor Arturo Casadevall, MD, of the Albert Einstein College of Medicine in New York, investigated retraction rates in 17 journals from 2001 to 2010 and came to the conclusion that retraction rates are rising at an alarming rate: 10-fold over the past decade, with steeper rates in higher-impact, leading medical journals. During that same decade the number of published papers had increased only 44%.2
All in the Dysfunctional Family?Part of this increase in retractions may be due to the fact that errors are more easily spotted now; journals are reaching wider audiences online, and digital searching makes it easier to come across articles that might have flown below the radar in the past.
However, Drs. Fang and Casadevall suggest the increase is a symptom of a “dysfunctional scientific climate” characterized by several factors tending to neutralize the normally self-corrective processes of rigorous science.
- Mounting pressure to publish as many papers and in as high a profile journal as possible
- Evermore cutthroat competition among PhDs for tenure-track jobs
- Once employed, pressure to continue to publish
- University promotions based on counting papers and grants rather than on the quality of research
The upshot, said Dr. Fang, is a sense that there is no time to check research and find out why some data do not fit the hypotheses (lest someone else publish similar results first). Dr. Casadevall expressed fear that perverse “winner-take-all” incentives among scientists breed corner-cutting and sometimes outright misconduct. The further long-term consequence, cited by Dr. Fang, is that scientists driven more by anxiety over funding than by a pursuit of knowledge are “not inspiring role models.”
Giving the journal’s perspective, JACC editor-in-chief Anthony N. DeMaria, MD, noted that “editors are poorly equipped to evaluate possible misconduct since we neither have access to the raw data of submissions nor the time and manpower to acquire and analyze it if we did.” In a JACC Editor’s Letter in April,3 Dr. DeMaria noted that the individuals in institutions in the best positions to identify research “misbehavior” are typically disinclined to sully their own reputations.
Dr. DeMaria cited a survey of 2,700 researchers that showed that 1 in 7 of the UK-based scientists or doctors had witnessed colleagues intentionally altering or fabricating data during their research or during the process of publication. Indeed, 2% of those surveyed (several dozen respondents) actually admitted to those behaviors.4
According to Dr. DeMaria, retractions at JACC have generally been initiated by the authors themselves after finding honest errors in their own work. But if, as reported, more than 10% of investigators are aware of scientific misconduct, then either the JACC review process has been “extraordinarily discerning,” or “we have occasionally been duped,” he acknowledged. In the final analysis, he concluded, “the validity of the medical literature has to be based upon the integrity of the community of investigators. Based upon my 35 years in academic medicine and research, I think that this confidence is reasonably well placed.”
That confidence was echoed by Spencer King, MD, JACC Cardiovascular Interventions editor-in-chief and president of St Joseph’s Heart and Vascular Institute in Atlanta. “There’s been no loosening of the journal’s scrutiny of papers. It’s as vigorous as we can make it at the present time.” He added, “I don’t find the fact that retractions happen to be an indictment. It means you are rigorous.” Dr. King said further that he does not know whether or not there are presently more people submitting papers with incorrect or fraudulent data than in the past.
Watchdog GrowlsIvan Oransky, MD, co-founder of the watchdog blog Retraction Watch (http://retractionwatch.wordpress.com) pointed out in an interview that last year’s 400 retractions (88 from one German researcher) is a huge chunk of the estimated 800 papers retracted over the prior 10 years. Admittedly, it is over a denominator of 1.4 million papers (0.03%), “so we can’t draw too many conclusions,” he said. He agreed, as well, that the greatly magnified number of “eyeballs” surveying online research today is a factor in the reported retraction increase. If true, then it is likely that not nearly enough papers were being retracted 10 years ago.
Neither Dr. Oransky, or Adam Marcus, his Retraction Watch co-founder, feels that an increase in misinformation and misconduct has been proved, nor the actual levers for it, if there are any, identified. However, he does believe there is a problem.
“How big it is, we don’t really know yet. But it certainly is bigger than ‘we don’t have to deal with it,’” Mr. Marcus said. He considers Drs. Fang and Casadevall to be “heroes” pushing hard for improved integrity and higher standards. He trusts their judgment, and believes that “everything they pointed out should be fixed anyway.” He added, “Whether or not that will fix the retraction problems remains to be seen.”
Dr. Oransky is deeply troubled by the calls he has fielded from scientists relating how incredibly difficult it is for them to get obviously flawed or even fraudulent papers retracted from the literature. “Sometimes we end up treating patients based on flawed research because editors find reasons not to clean up the literature,” he said.
Commenting on Dr. DeMaria’s JACC letter, Dr. Oransky said, “I didn’t find it reassuring when he said that journal editors are not equipped to evaluate possible misconduct. If you want people to believe that what’s in your journal has been vetted and that any questions that have arisen have been properly dealt with, then you should have the resources to do that. If you are publishing more papers than you have time to vet, that’s a problem. That’s throwing up your hands.”
Dr. DeMaria told CardioSource WorldNews that he agreed that any questions of misconduct that arise should be properly investigated, and indicated that this is the existing practice. However, he added, authors are assumed to be honest, and routinely evaluating raw data over and above representative illustrations is most effectively done by the home institution of the authors.
What solutions are being proposed? Only general directions seem to be indicated. Drs. Fang and Casadevall say that the overall climate in science is darkened by winner-take-all grant-giving, and that lavishing all discovery credit to first publishers fosters secrecy. Similarly, the scientific virtue of collaboration is ignored in favor of prizes and promotions. Dr. Oransky said, “We should stop fetishizing the scientific paper, and we should be willing to have a much more robust post-publication peer review system. We need to be able to continue to trust the so-called self-correcting nature of science.”
Whether or not soul-searching on these themes will effect change remains to be seen.—by Walter Alexander
*The last seven retracted papers by Naoki Mori of the University of the Ryukyus, Japan, had already been cited 106 times.
- Zimmer C. “A Sharp Rise in Retractions Prompts Calls for Reform.” The New York Times. April 17, 2012; D1.
- Van Noorden R. Nature. 2011;478(7367):26-8.
- DeMaria AN. J Am Coll Cardiol. 2012;59(16):1488-9. http://content.onlinejacc.org/cgi/content/full/59/16/1488
- Tavare A. BMJ. 2012;344:e377.
Editor’s Note: Opinions expressed in the preceding article by Anthony DeMaria and other sources represent the opinions of those contributors, and are not necessarily the opinions or policies of the ACC, of the ACC Foundation or of the editors of the JACC family of journals.
< Back to Listings