A 'Disappeared' Article, Finally Published, Finds A Desultory Response To Scientific Misconduct Back in January an article in the European Heart Journal raising the disturbing possibility that guidelines based on fraudulent research may have been responsible for as many as 800,000 deaths was Medicine Or Mass Murder? Guideline Based on Discredited Research May Have Caused 800,000 Deaths In Europe Over The Last 5 Years from the journal’s website only minutes after being published. The journal’s editor claimed the article had not been properly peer reviewed. In its place the editors published an editorial refuting the claims of the article, though of course they were responding to an article that no one could then actually read. Now, eight months later, a revised version of the original article, by British cardiologists and gadflies Graham Cole and Darrel Francis, has finally been published. The new version raises even more disturbing questions about the potentially lethal effects of such fraudulent research not only on clinical practice but on the credibility and worth of the entire scientific enterprise.
The original article by Cole and Francis explored the consequences of the damage caused by the Don Poldermans research misconduct case. Poldermans research provided the bulk of the evidence to support the liberal use of beta-blockers to protect the heart during surgery for people undergoing non cardiac surgery in the European Society of Cardiology guidelines. Cole and Francis calculated that this guideline, based on questionable and probably fraudulent research, may have been responsible for as many as 800,000 deaths over 5 years in Europe. In the new revised version of the article Cole and Francis provide a detailed dissection of the new updated guideline. They then focus on the broader implications of the scandal. They write about a culture of neglect in which few if any participants have anything to gain by finding or reporting scientific misconduct. They cite numerous examples in which misconduct has been alleged but the responsible actors– authors, home institutions, journals, and medical socieites– have responded in only the most minimal and nonaggressive fashion. The portrait they paint is of a scientific and medical establishment devoted to not rocking the boat. Francis and Cole lament the harm that can occur by an entrenched medical establishment. Poldermans, as the chair of the discredited guideline, was hardly likely to undertake a critical perspective of his own trial. Perhaps even more disturbing, his co-author and frequent collaborator, Jeroen Bax, was the head of the entire ESC guideline system. (The ESC has announced that Bax will be the next president of the ESC. Bax’s role in the Poldermans affair has never been publicly explained.) Francis and Cole offer a number of specific observations and proposals. One strong recommendation is for co-authors of papers (Bax, for example) to assume more responsibility and to play a more vigilant role. ”Co-authorship,” they write, “brings status and metrics but no harm when the study is found misconducted.” But co-authors may be uniquely able to shed light on misconduct. “If we made them all share the consequences when research is misconducted, they would try harder to prevent it.” Journals also need to play a far more important and aggressive role in detecting and handling scientific misconduct. “Journals informed of internal contradictions should ask the authors to immediately provide the full raw data set so that readers can verify the source of the error. If authors delay, journals could immediately temporarily retract the article…” Another key recommendation, in line with the AllTrials.Net and Yale Open Data Access Project, is that “authors should try to publish ‘all the data, all the time’.” Openness, they note, “protects all of us from any temptation to edit data to match expectations.” Cole and Francis also discuss the central importance of community responsibility: If criminals claim ignorance of information that would make them culpable (e.g. ‘I did not ask what was in mysterious packages I was handsomely paid to transport’), courts may recognize this as ‘wilful blindness’ and thereby convict them. As doctors we should never give in to the temptation to fail to notice the obvious: whether as co-workers reading trials that we know did not happen in our hospital, co-authors offered credit on misleading manuscripts, readers noticing factual impossibilities, or editors told of serious problems. Rather than being a badge of shame, in some cases retractions should be a mark of honor and honesty: “We should be generous when authors who, having presented misleading reports, later correct them. Everyone makes mistakes: good scientists look for and proactively fix their own errors…. Shame and ostracism should be reserved for authors (or others) in denial, obstructing resolution, or hiding misdeeds through legal threats or censorship.” Finally, in accord with Cole and Francis, it is probably worth pointing out that it took several years for the ESC to change its guideline following the initial accusations leveled against Poldermans, but it took several hours for the EHJ editors to remove the original version of the Cole and Francis article from the EHJ website. Perhaps one day the same sense of urgency and action will be devoted to scientific misconduct as it has been in the past devoted to protecting reputations and the status quo.
