A hard-hitting analysis in the June 19 British Medical Journal questions the reliability of clinical practice guidelines, suggesting that in many cases they have become more industry marketing than tools of evidence-based medicine. (2013;346:f3830.)
After touching on the recent policy change from the American Association of Neurological Surgeons and the Congress of Neurological Surgeons recommending against steroids in early spinal cord injury, a declaration many considered overdue, investigative reporter Jeanne Lenzer turned to an issue that has long pitted neurologists against emergency physicians: tissue plasminogen activator (tPA) for treating acute ischemic stroke.
New joint guidelines from the American College of Emergency Physicians and the American Academy of Neurology were issued on tPA this past spring (see FastLinks), endorsing the FDA-approved three-hour treatment window for tPA as well as a three-to-4.5-hour window in “carefully selected patients.” Ms. Lenzer's article brought to light undisclosed conflicts of interest among some members of the guideline panel.
“In the guidelines ... three of eight panelists disclosed ties to the manufacturers,” she wrote. “However, seven had either direct ties to the manufacturer or indirect ties, knowingly or not, through affiliations with the Foundation for Education and Research in Neurological Emergencies (FERNE), which provides unrestricted continuing medical education grants. Guideline readers were unlikely to know that according to its 2008 tax return, 100% of the $97,000 donated to the foundation that year came from drug companies, including $50,000 from Genentech. The foundation president and founder, Edward P Sloan, is an outspoken advocate of alteplase for stroke.”
Ms. Lenzer noted that industry influence on clinical practice guidelines isn't unusual, pointing to a 2012 analysis in the Archives of Internal Medicine that reviewed 130 randomly selected guidelines from the National Guideline Clearinghouse. The authors found that conflicts of interest were present in more than two-thirds of committee chairs (71.4%) and 90.5 percent of co-chairs.
And as Emergency Medicine News reported last year, 85 percent of the most cited articles in the tPA literature were found to be tied to pharmaceutical interests, according to an analysis published in the Western Journal of Emergency Medicine. (See FastLinks.)
“All I can say is that, speaking for myself, I have no conflict apart from what I declared,” said the guidelines' lead author, Jonathan Edlow, MD, the vice chair of emergency medicine at Boston's Beth Israel Deaconess Medical Center. Dr. Edlow did disclose his ties to FERNE in the published guidelines. He was an unpaid board member, and has given lectures for FERNE on various neurological subjects but never on acute ischemic stroke or tPA. Dr. Edlow also noted in his disclosure that he testifies in legal cases involving stroke for the plaintiff and defense. He said 90 percent of cases in which he had an opinion were in support of emergency physicians for not giving tPA.
Ms. Lenzer also questioned “committee stacking,” or loading down panels with experts with known viewpoints on a particular side of an issue. She wrote that seven of eight members of the AAN-ACEP panel had previously spoken or written on the benefits of tPA for stroke. “Not one skeptic was included on the panel,” she wrote.
One listed committee member, Robert Wears, MD, PhD, a professor of emergency medicine at the University of Florida Health Science Center and the group's methodologist, had actually resigned from the panel some years previously. Mrs. Lenzer wrote that he had become “disillusioned” with the guideline process, and Dr. Wears told EMN that he did not resign out of disappointment over the way things were going with those guidelines specifically or in objection to conflicts of interest.
“I left clinical policies because I'm not convinced that they're ever going to be useful, except perhaps in the context of education,” said Dr. Wears, a self-described “agnostic” on tPA. “They may be helpful to a novice as a way to get started, but in terms of fulfilling the original vision for them, they're inherently flawed. They are thin rationalizations of very complex issues. Clinicians don't really adopt them and follow them. If the customer isn't buying what you're selling, maybe there's a problem with what you're selling.”
Dr. Wears said Ms. Lenzer also raised a very legitimate issue with the problem of industry funding and bias infecting the guideline process, but he dismissed the notion that it's a grand conspiracy. “It comes across as, ‘Let's get together because it'll make a lot of money for us.’ I don't think that's the case, particularly here. The big problem with all kinds of research is that the people most interested in a topic tend to be the ones who've gotten funded to do it, and some of that funding either comes from industry or is connected to it.”
He conceded that the AAN-ACEP panel was drastically out of balance, but said the panel probably would not have reached consensus at all with ardent opponents of tPA represented. “You could argue that this might have been a reasonable result, given the degree of disagreement that is present in the field. It may have been premature to come out with global recommendations one way or the other, particularly in a cross-specialty-nationwide sense,” Dr. Wears said. “It requires careful consideration of local factors, local capabilities, and willingness to accept risk on either side for an organization to decide how they're going to approach the tPA question.”
The two other tPA panelists from emergency medicine were Andrew Jagoda, MD, a professor and chair of emergency medicine at New York's Mt. Sinai Hospital, who Ms. Lenzer reported had disclosed his ties to FERNE but had not revealed his relationships with Genentech and AstraZeneca, and Latha Ganti Stead, MD, the chief of clinical research in emergency medicine at the University of Florida College of Medicine, who had no conflicts. Dr. Jagoda declined to speak to Emergency Medicine News, and Dr. Stead did not respond to inquiries.
Dr. Wears suggested that seeking more rigor is not the way to go on a question with a body of evidence as conflicted and controversial as tPA. “That's the mistake that clinical policy people are making. The more rigorous you get, the less relevant the results become. People who manage to get enrolled in clinical trials do better than normal people even in the control group,” he said. “And you can have very rigorous evidence that is not very compelling for a variety of reasons, and evidence that is not so rigorous that is nonetheless very compelling because of factors such as biological plausibility, balance of risks and benefits, and ability to isolate subgroups of patients who will benefit.”
Researchers and guideline writers should look more at practical implementation instead of pitting treatments against each other and the holy grail of the p-value, he said. “Can you really do this successfully on a 24-7 basis? In whom does it work, and at what costs?”
The tPA debate is unlikely to be resolved any time soon. “It's gotten so polarized: the people who do not think that tPA is a proven entity act like tPA is poison and the tPA believers act like it's holy water, and it's neither,” said Dr. Edlow. “It's a treatment that helps some patients. If you take 100 eligible patients, in a health care setting that is properly set up to give it, there will be about 11 or 12 who do distinctly better and four or five who do distinctly worse, but 85 percent won't do any better or worse because of it.”
Click and Connect!Access the links in EMN by reading this issue on our website or in our iPad app, both available onwww.EM-News.com.
- Read EMN's article about the AANS and CNS policy change recommending against steroids in early spinal cord injury at http://bit.ly/18YbSWt.
- The EMN article about the AAN and ACEP tPA guidelines is available at http://bit.ly/EMN12.
- Read Dr. Luis Lovato's EMN column, “Pharma Bias Detected in Thrombolytic Stroke Trials,” at http://bit.ly/13WsMGN.
- Comments about this article? Write to EMN at email@example.com.