The Politics Behind the Fall of Avandia
How one doctor and statistician took down the number one diabetes drug.
Let's open with a disclosure!
In trying to translate complicated diabetes messages in a light, humorous tone, I typically shy away from the underpinnings of politically charged issues such as the Avandia debacle. After observing the ongoing debate from behind the scenes and it's untold impact on the diabetes community, I feel it is my duty to share an inside perspective so that readers have the information to make an informed choice. For more background to this article, please read, "To TZD or not to TZD". These are not the views of dLife. I have never received a cent from GlaxoSmithKline (the makers of Avandia), nor is this an endorsement for using it.
It's a classic case of benefits versus risks, and unknown political motivations.
After Dr. Steve Nissen and statistician Kathy Wolski M.P.H. co-authored, "Effect of Rosiglitazone [Avandia] on the Risk of Myocardial Infarction and Death from Cardiovascular Causes," published earlier this year in the New England Journal of Medicine (NEJM), the ensuing media frenzy put the diabetes community into a tailspin.
Nissen and Wolski did a meta-analysis, a statistical technique that compares outcomes of selected studies. This approach in itself has flaws. If you "cherry pick" studies that show less then favorable outcomes, the results of the meta-analysis will reflect negatively as well. Furthermore, the studies compared in the Nissen / Wolski study were not designed to look for adverse heart-related events (data was in some cases self-reported and could not be verified). The media went hog wild and sensationalized the skewed results. Many people panicked (including those in healthcare), and removed Avandia from their diabetes toolbox.
Patients and their healthcare providers began to immediately limit their options by automatically changing Avandia to Actos, or canning the TZD group of medications altogether, a class of drugs proven to reduce insulin resistance (IR). IR often leads to increased blood pressure and cholesterol, known factors causing heart disease, the very issue taken up by Nissen and Wolski.
Who are Nissen and Wolski? Dr. Steve Nissen is the Chairman of the Cleveland Clinic's Department of Cardiovascular Medicine where Kathy Wolski is employed as the lead biostatistician. Dr. Nissen is also the Past-President of the American College of Cardiology and world-renowned cardiologist, named one of the 100 Most Influential People in the world by Time magazine.
Let me try to explain their method of analysis by using the following example:
Study A had 15,560 Fuji apples (86 were bruised, 39 weren't any good).
Study B had 12,283 Washington apples (72 were bruised, 22 weren't any good).
Through statistical analysis, when comparing the two groups, we show a 43% increase in bruised apples in study A (or 14 apples) and a 64% increase in apples in study B that weren't any good (or 17 apples). Not that people are apples! This illustration shows that a relatively few numbers can have dramatic consequences when you start comparing the data in percentages.
What about the cup-half-full notion? Why not say that in the selected meta-analysis study with Avandia, only 39 out of 15,560 people died and 86 had a heart attack, whereas the smaller group without Avandia, only 22 out of 12,283 people died and 72 had heart attacks? Please put this into perspective. Roughly 60% of people with uncontrolled diabetes die of heart problems. This flawed meta-analysis backs up what we already know about uncontrolled diabetes and heart consequences.
Keep in mind the average A1C for the short study duration of 24 weeks was 8.2%. We have no idea about the prescribing practices of individuals who received the drug (was it safe for them to receive it?), or a head-to-head comparative status of their cholesterol, blood pressure, aspirin use or stress levels, all components necessary for optimal heart health. Further, 50 % of people have some type of complication when first diagnosed (that's not Avandia's fault – it's lack of early diagnosis). And if that's not enough to chap your hide, as a nation, we are not fairing much better with glucose control in the past decade (the mean A1C in 1990 was 7.8%, and in 2000 it was 7.7%).
Nissen uses heart attacks as end points and argues we should look at reduction of diabetes-related complication rates in addition to glucose levels. Then he needs to pull from studies that impact cardiovascular health, including cholesterol, blood pressure, and aspirin use, before running to tell Congress and the FDA. Plus, the FDA has stated that to date, no oral anti-diabetes drug has been conclusively shown to reduce cardiovascular risk. So what is Nissen's beef? He wants a drug to do something it wasn't set out to do? Avandia is for insulin resistance; it's not advertised to decrease cardiovascular issues related to uncontrolled diabetes.
If you research Nissen on the web, you will find reputable journals describe him as an alarmist, political, an FDA watchdog ("angling to become the next FDA commissioner"), self-promoter, someone who released premature analysis, and other unflattering descriptions (like the "Michael Moore of the medical industry"). He acknowledges being in communication with Congress ahead of publication of the meta-analysis in NEJM. He is a self-described political leader whose motivations are in now question. Nissen was instrumental in getting Vioxx and Pargluvia off the market. What's next? If he does step into the FDA arena, how will that impact the diabetes community and the pending generation of diabetes medications? He claims not to have interest.
It seems odd from my perspective as a certified diabetes educator, why a cardiologist would want to be the mouth-piece for medication safety, have inroads to members of Congress, and practically single-handedly bring several mainstream drugs to their knees without having political or career intentions. Then again, I love a good conspiracy.
During a CNBC interview, Nissen explained that the alternative is not to know and that not to publish is not acceptable. "My job is a scientist," said Nissen, who added, "When you see something like this, you publish it…It's not about politics, it's about patients." Nissen said, "The truth is what counts. You can only understand the truth if you have all the information." In my opinion, which is shared by many in the diabetes scientific community, he didn't have all the information needed to present unbiased findings.
During a special symposium that had to be added to the American Diabetes Association annual conference in Chicago (4,500 attended in a standing room only), Dr. Nissen apologized for making "life more difficult" for the many diabetologists in attendance.
My hope is that your diabetes life is not more difficult as a result.
It is not too late to have a candid conversation with your advocates about your individual benefit or risk of using Avandia (or Actos). Healthcare decisions should not be made on the advice of a self-proclaimed outsider, but between you and your trusted healthcare team — the people who have the most information about your health situation.
Read Theresa's bio here.
Read more of Theresa Garnero's columns.
NOTE: The information is not intended to be a replacement or substitute for consultation with a qualified medical professional or for professional medical advice related to diabetes or another medical condition. Please contact your physician or medical professional with any questions and concerns about your medical condition.
Ginger Glazed Salmon Simple Spaghetti Sauce Fruit Medley Salad Cabbage and Potato Soup with Caraway Chayote Squash Soup with Cilantro Sour Cream Enchilada Sauce Spinach-Mushroom Saute Beef Burgundy for the Slow Cooker Creamy Banana Topping Turkey Breast with Barley-Cranberry Stuffing
As I mentioned in an earlier post, one of the benefits that made it cost-effective for me to go with the real healthcare (HSA) plan rather than the phony (HRA) plan is that my company is now covering "preventative" medicines at $0 copay. The formulary for these, as stated by CVS/Caremark (my pharmacy benefits provider), covers all test strips, lancets, and control solutions. I dutifully get my doctor to write up prescriptions for all of my testing needs, submit...