Monday, December 22, 2008

CT Angiography, Dr. Nissen and Conflicts of Interest

For the record, I am not in favor of CT angiography, primarily due to the large doses of radiation. EBCT (Heartscans) have yet to be proven, but seem a much better technology. Last week the Wall Street Journal (via KevinMD) reported on a study in the Journal of the American College of Cardiology looking at the diagnositic accuracy of this test. Though the investigators (the lead author being a Netherland's radiologist) stated that CT angiography was " reliable for ruling out significant CAD in patients with stable and unstable anginal syndromes," what was reported in the media was Dr. Steve Nissen's comment and editorial that "in more than 50% of subjects, CT angiography ‘detected’ coronary obstructions that simply were not there.” Dr. Nissen's comments are not incorrect, and I do agree with his position on this one, but it should come as no surprise that a cardiologist (especially one's whose research primarily involves conventional angiography) comes out against CT angiography.

Conflicts of Interest
Conflicts of Interest in the medical field as reported in blogs and in the media usually concern the pharmaceutical industry. There are plenty of folks out there (Dr. Carlat and Center for Science in the Public Interest to name just a few) who bring attention to this. There is no question that a drug study sponsored by the drug's maker should be interpreted with caution. However, this doesn't mean that the data should be ignored. For example, when the Toyota car salesman tells you that the Prius is one of the most fuel efficient and reliable cars, he is certainly trying to get you to buy a Prius, but it doesn't mean his data is wrong or that you shouldn't buy the Prius. You just need to interpret his "data" with some caution. This concept does not just apply to the pharmaceutical industry, but other experts or groups of experts that make recommendations that may be biased based on other factors such as research, research interests, specialty, and even political motivation.

Conflicts in Guidelines
Guidelines are the best examples when conflicts of interest should be taken into account. The American Urologic Association currently recommends PSA testing for prostate cancer in men 50 and over. The U.S. Preventive Services Task Force not only states that "the current evidence is insufficient to assess the balance of benefits and harms of prostate cancer screening in men younger than age 75 years" but also recently recommended that prostate cancer NOT be screened for in men age 75 years or older. Both guidelines are written by medical experts, and both have looked at the same studies. The difference in interpretation has to do with potential biases. Urologists who see the horrors of prostate cancer on a regular basis, likely want to eliminate this disease, and additional have a financial incentive to detect prostate cancer (the more PSA tests, the more biopsies, the more surgeries). In contrast, the USPSTF is a government sponsored agency concerned not only with the health of Americans but also the cost. If PSA screening leads to more testing, biopsies and surgeries; then this will be more costly for the health care system (especially Medicare). Thus, neither interpretation of the same data is "right" or "wrong" but must be interpreted with attention to potential biases or motivations.

Other Conflicts
Research interests are another potential conflict. Someone who researches in a particular area is not likely to discredit what has made their career. Recently, I blogged about the JUPITER study which showed that CRP may be a very important marker for cardiovascular risk. It should be pointed out that Dr. Paul M Ridker, the lead author of the study, has made his career on CRP research and currently holds a patent on the test used in this study. Political motivation can also be a factor. In my comments on the fall out of the recent FDA panel's meeting on the safety of LABA's, I mentioned that Dr. David Graham had commented against these drugs (picked up by the NY times), but that he may have been politically motivated for making these statements since he was mostly wrong back in 2004 when warning congress of the "next Vioxx" and, as part of the safety arm of the FDA, alerting the public to safety issues might increase funding to his (agreeably) underfunded division in that agency. Which brings me back to Dr. Nissen.

Dr. Nissen's Potential Conflicts
Dr. Nissen is a cardiologist at the Cleveland Clinic, which recently announced that it would make its faculty's conflict of interest public. Interestingly,they report that Dr. Nissen has no conflicts, and though he does consult for a number of drug companies, he donated "all honoraria or consulting fees directly to non-profit organizations." First, no conflict is reported because the Cleveland Clinic does not consider research funding as a conflict of interest. In fact, Dr. Nissen, who has come out strong against Avandia, has had publications and research supported by Takeda, which makes Actos, Avandia's major competitor. In addition, the anti-pharma folks (correctly) suggest that any gift or relationship can pose a potential conflict. I do not criticize Dr. Nissen for his likely excellent consulting work, and would not do so even if he kept the money (since an expert should be paid for his time and expertise), but donating the money does not exempt him from any conflict. Finaly, and possibly most importantly, Dr. Nissen gained fame (which he deserves) for bringing attention to Vioxx, and helped get it removed from the market. Because of this attention (and the attention from Avandia) Dr. Nissen has been mentioned as one of the candidate to head up the FDA. Thus, until President Obama picks the person for the job, one should expect (as in the case of CT angiogram) that any comments from Dr. Nissen will likely reflect his desire to protect the public from unnecessary testing or the dangerous medicines from profit hungry drug companies.

My beef with Dr. Nissen is not that his research is funded by drug companies or that he feels donating honorarium exonerates him from conflicts of interest. Rather, as I have blogged many times (here, here here and here), his one poorly done study which scared the public and physicians about Avandia, has radically changed the way type 2 diabetes is being treated (and not for the good of patients in my opinion) and no one is calling him out on potential conflicts or motivations.


james gaulte said...

Your entries get better and better.One reason for me to get blogging again is to reference your stuff.

Anonymous said...

With Nissen- note that his contributions to charity go to his own chartitable fund. What happens to those dollars? I dont doubt that he does donate them to worthy causes - I just imagine it becomes his own little political leverage tool.

Also - check out Nissen's similar metaanalysis for muraglitazar, a dual PPAR diabetes drug that was basically killed because of the same shoddy meta analysis techniques. He actually killed more than just that drug - he killed the whole class.

Unknown said...

OMG, as much as I supported Obama, if her were to make the mistake of appointing this fraud as FDA director, I think I would leave medicine. What a joke that would be!