The Globe's Stephen Smith discussed the process leading to the publication of the Screening for Heart Attack Prevention and Education (SHAPE) Task Force guideline on screening for coronary artery disease in the American Journal of Cardiology (AJC). (Full citation: Naghavi M, Falk E, Hecht HS et al. From vulnerable plaque to vulnerable patient - part III: executive summary of the Screening for Heart Attack Prevention and Education (SHAPE) Task Force Report. Am J Cardiol 2006; 98[suppl]:2H-15H.)
The SHAPE Task Force is apparently a group of cardiologists who strongly advocate high-technology screening to detect asymptomatic coronary artery disease (CAD). The Task Force wrote a guideline to this effect, even though, per the Globe, "the authors concede [the guideline] is not supported by rock-solid evidence." Unable to get official support from the American Heart Association or the American College of Cardiology, they asked the AJC to publish it in a special supplement. So,
Journal editor Dr. William C. Roberts told the group that, in contrast to how it works in the regular pages of the magazine, if they wanted their recommendations published, they 'would have to have some financial support.'Once the authors came up with the funding from Pfizer, which summed to $55,800 according to the Globe, the AJC was willing to publish the guideline, which was thereafter "not subject to the standard review process."
Dr. Morteza Naghavi, lead author of the guidelines, sent letters soliciting aid to six drug companies. In the letters, which Naghavi supplied to the Globe, he writes that 'the report will be distributed to 100,000 physicians worldwide.'
Naghavi received a favorable response from Pfizer Inc, whose funding of the report is noted in the journal. In an e-mailed answer to an inquiry from The Globe about its contribution, a company spokeswoman wrote that 'Pfizer feels it is important to provide support for efforts that assess novel approaches to reduce the burden of heart disease.'
According to the Globe, Pfizer's "funding of the report is noted in the journal." Furthermore, "in an appendix to the heart-screening guidelines, several authors acknowledged that they had financial arrangements or affiliations with drug companies or medical device-makers whose products might be influenced by these recommendations."
The guideline is available on the web here, along with an introduction by its authors. As best as I can tell, the versions available on the web do not include any mention of the funding by Pfizer, the financial arrangements of the authors noted above, or any notice that the guidelines were not subject to the standard peer-review process. Several of the guideline authors were listed as Pfizer employees. (ADDENDUM, 24 August, 2006: A comment below suggests that the Pfizer funding is disclosed in the print edition of the journal. I do not have access to the print version, so cannot confirm this, but have no reason to doubt it.)
The guidelines themselves advocated the use of the coronary artery calcification score (CACS) determined by computed tomography (CT) scanning, and the carotid [artery] intima-media thickness (CIMT) determined by ultrasound as screening tests to detect CAD. Although the guidelines were based on extensive citations in the literature, they were not developed using an explicit evidence based medicine process. In particular, the authors did not systematically survey all the relevant literature, and did not critically review the most pertinent articles.
The criteria the guidelines used for recommending screening tests were
(1) the abundance of evidence for the predictive value of the test in the recommendedThese criteria were markedly different from the criteria used by the conservative and evidence-based US Preventive Services Task Force in making screening recommendations. For example, the SHAPE Task Force guidelines seemed to ignore the sensitivity of the tests, whether there was evidence that treating asymptomatic patients whose disease was diagnosed by screening tests would provide better results than waiting to treat patients when they developed symptoms (and particularly whether aggressive treatments of large numbers of asymptomatic patients, particularly with lipid-lowering drugs, might have some adverse effects), and whether the tests, again when employed widely in a screening application, could have adverse effects (especially given the radiation dose from CT scans.)
population over and above that available from standard office-based risk assessment tools (incremental value), (2) availability, (3) reproducibility, (4) complementary
value with respect to the concept of the vulnerable patient, and/or (5) cost-effectiveness relative to the status quo.
As former New England Journal of Medicine Editor Dr Jerome Kassirer put it in the Boston Globe,"the whole thing sounds like a conflicted mess, from the recommendations that they're making to the issue of how these journal supplements work." Furthermore, not all the conflicts were disclosed, at least in the web-based version of the guidelines. On the web, the journal failed to disclose that this supplement was published without peer review and was supported by a pharmaceutical company, and that some of the authors had potential conflicts of interest.
Again, this case suggests that physicians and patients need to fully understand the financial (and other) arrangements that could affect opinion and research articles published in medical journals, even when the articles have distinguished authors and are published in reputable journals. We have now seen all manner of such arrangements affecting articles' authors, their institutions, and the journals that publish them. Many of them have gone totally undisclosed, unless discovered by enterprising journalists or academic whistle-blowers. Even those that have been disclosed have not been disclosed in sufficient detail to really evaluate their effects.
Physicians and patients now need to be extremely skeptical about whether such arrangements affect the content or arguments of seemingly scholarly articles published in even the most reputable journals, and may need to be more skeptical about articles which disclose no such arrangements than about those which at least offer some partial disclosure.