How to survive the health care system.

You have heard about good and bad cholesterol. You have heard that increasing the former and reducing the latter will cut your risk of heart disease. You will now hear what's principally wrong with this strategy of attacking risk factors. And how it prevents us from eradicating the heart disease epidemic sweeping the globe. 

On 30th November 2006, Jeff Kindler, the CEO of Pfizer, praised their about-to-be released drug for increasing good cholesterol as "...one of the most important compounds of our generation."

Three days later Pfizer halted the phase 3 clinical trial of its hoped-for blockbuster drug. The simple reason: the drug's ingredient, torcetrapib, did what it was supposed to: increase good cholesterol. But it also increased patients' risk for heart attack, stroke or death from any cause [1]. You probably see the common theme behind this and the story of my previous post: a drug improves a risk factor but worsens risk. Fortunately, Pfizer didn't cover that up.

Now, before you admire Pfizer as an outstanding citizen of the pharma world, let's look at their track record in the ethics department: In 2009 Pfizer pleaded guilty to a felony violation of the Food, Drug and Cosmetic Act for misbranding their drug Bextra and three others "with the intent to defraud or mislead".

The anti-inflammatory drug Bextra had been pulled off the market 4 years earlier, but Pfizer bribed its way into physicians' prescription blocks. The consequence: a $ 2.3 Billion criminal fine, the largest ever awarded. You would think of such a fine as putting a serious dent into a company's balance sheet. Well, 4th quarter profits 2008 were only 10% of what they used to be. Pfizer had made a provision for what they knew was coming. But compared to the $ 50 Billion in annual sales that's nothing over which a CEO would lose sleep.

Last month, Merck was fined $ 321 Million for similar offences related to Vioxx, a drug of the same class as Bextra.

Altogether, Big Pharma has been fined $ 8 Billion over the past 10 years for repeatedly defrauding the U.S. health care system. If they do it repeatedly, it's probably not because they are slow learners. It's because - 'cherchez l'argent' - there is money in it. If that's the case, then what has been brought into the open, may just be the proverbial tip of the iceberg. Now, an iceberg of monetary fraud is one thing, an iceberg of defrauding you of your health is quite another. Let me backup my suspicions, again with a recent example: Tamiflu

Roche's Tamiflu is the only orally administered influenza antiviral drug in its class (neuraminidase inhibitors). Governments all over the world had been stockpiling it before the 2009 influenza outbreak. So did the U.S. government at a cost of $ 1.5 Billion for Tamiflu and Relenza. 

Based on what evidence? Chiefly on a meta-analysis of 10 studies which, in 2003, informed the public that Tamiflu substantially reduced the need for antibiotics and also reduced the risk of developing serious complications [2]. What's wrong with that? That 8 of the 10 studies cited in this Roche-sponsored meta-analysis had not been published in peer-reviewed journals. That is, independent researchers have been unable to verify such claims. Not that they didn't try. The most authoritative organization for conducting meta-analyses is the independent Cochrane Collaboration. The Cochrane researchers Peter Doshi and colleagues attempted to verify the claims of the Kaiser meta-analysis, so named after its lead author. But despite repeated requests, Roche has remained uncooperative in sharing the full data reports, offering reasons of which Doshi and colleagues had to say that "none seemed credible" [3].

In a March 2012 interview with the Swiss newspaper Neue Zürcher Zeitung, the head of the German arm of the Cochrane Collaboration, Prof. Gerd Antes, enlightened readers about Roche's way of doctoring the Kaiser meta-analysis. Not only did Kaiser NOT have access to the 8 studies, Roche had simply given him their evaluation of these Roche sponsored studies' data. And the other two studies alone do NOT support the overall "result" of the meta-analysis. If it wasn't OUR tax dollars and OUR health, I'd say it was funny how governments all over the world are gullible (that's the polite version for 'stupid and reckless') enough to shovel billions of Dollars and Euros into Big Pharma's pockets. Based on nothing else than doctored data. With my small lab we are currently jumping through hoops to get a few ten thousand Euros from a government research fund for a project that will take us 2 years to accomplish. I'm not complaining, I just mention it for contrast.

If I was a conspiracy buff, which I am really not, I would suspect a put-up affair: Politicians first accept lobbyists' contributions for which in return they waste our money on useless, or even dangerous, drugs, after which they earn brownie points from us by publicly wrist-slapping Big Pharma for their deception with fines which look big but really are not. With Pfizer alone spending annually around $ 12 Million for lobbying to lawmakers, Roche and Merck  each $ 8 Million, this scenario doesn't look that far fetched. At least for Big Pharma the return on (lobbying-)investment appears obscenely good. 

Anyway, what you have seen in the case of Tamiflu is called publication bias. It means that positive results are more often reported and published than negative ones. This bias comes with a serious side effect: Meta-analyses and reviews, which typically inform medical practice guidelines, can only be as accurate as their knowledge base. If that base is biased, so will be the treatment or advice which you receive from your doctor.

How can we gauge the extent of this problem? That's surprisingly simple, provided you have access to files of the ethics boards and committees. Every study must be approved by an ethics committee before it can be carried out. Prof. Antes has this type of access. He and his Cochrane team found that of all studies, which received an ethics committee's approval since 2000, only 50% had their results published. Which simply means for every published study there is another one, which, in all likelihood, comes to a different conclusion with respect to effect and risk of its subject drug or treatment.

That is frightening. You might wonder how medicine can be such an inaccurate science. It's because we still know so little. Every biomarker, every hormone every molecule represents one small node in an immensely intricate network of biochemical pathways. At each node sebverlas pathways may intersect, some of them we know, many we don't. So when we develop a drug which manipulates one pathway through one of these nodes, we inadvertently interfere with other pathways. These effects may or may not turn up in the three phases of clinical trials. Recall the torcetrapib example, which I mentioned in the beginning of this post, and the glitazone example of the previous post.

That's why reducing risk is not as straight forward as reducing risk factors. Take the all-time favorite, high blood cholesterol. Almost two thirds of newly diagnosed Indian patients with coronary artery disease (CAD) have normal blood cholesterol levels [4]. Then there is the mother of all risk scores, the Framingham Risk Score (FRS). It uses age, gender, total cholesterol, good cholesterol, blood pressure, smoking and diabetes status to compute your 10-year risk for heart disease. But more than 40% of those people who develop the disease, are flying below the FRS radar. That is, they wouldn't have raised a red flag even if their doctor had examined them the day before their heart attack [5].

Now, you probably think that we are making progress in identifying more and better biomarkers to predict disease risk. Boy, are you wrong. In a review, published just 2 months ago, its authors rounded up 36 new biomarkers which are not already included in the FRS [6]. Of those 36, they investigated the 10 most promising, as determined by the numbers of studies published. Together these 10 had almost 123,000 studies to their names. Mind you, that's only the number of studies with a focus on cardiovascular disease! To what effect? None for most of them. And for the three with a moderate predictive value, BNP, CRP and fibrinogen, the authors found evidence for, you guessed it, bias.

Performing studies is no cheap business. So why burn money on biomarkers, which we know to be of little use?  Because they have great monetary value. Think about it. Once a biomarker, such as cholesterol, has been accepted as a risk predictor into clinical practice, developing drugs which improve that biomarker is the logical next step. And once the widespread use of the cholesterol lowering drug is found to correlate with some reduction of heart attacks and strokes, public health is happy with this drug. Because in public health, we are not concerned with you or with your personal health. We are concerned with the health of the population at large. Back to our example of the cholesterol lowering drug.  The more people qualify for its prescription, the larger the business of developing, producing and distributing this drug. Voila, here is Big Pharma's big business.

Now, let me ask you a hypothetical question: What would you be prepared to do to prevent a $ 1.8 Billion investment from going down the drain? That's the current cost estimate for bringing one drug to market [7]. It takes into account that only a fraction of the molecules, which show some promise in laboratory rats, will make it into a phase 1 clinical trial where they are tested on humans. And only one in 5 of those which make it to phase 1 trials will make it all the way to phase 3 and from there to your pharmacist. With more than 10 years for the entire process, mind you. So let me refine my question, what would you do with your drug which, in phase 3 trial, shows some effect but also a side effect in quite a number of participants? You would certainly NOT be tempted to hide these effects from the public, because you are such a nice person. But some other people, who also have shareholders  breathing down their necks, well, they might be tempted to swing the trial data in favor of the expected blockbuster earnings, don't you think?

Which brings us to the question: how can you survive this system with your health intact? The simple answer is, don't develop any risk factors in the first place. Stay out of this health care system. At least don't find yourself in it as a patient with a chronic condition. What that means to your chances at longevity and celebrating your 80th birthday in full health, you have read in my previous post  "When risk scores for heart attack really suck!"

In the next post I will tell you what your individual way to optimal health and longevity will look like. And how we, in my lab, work on making an individualized prevention strategy a reality, soon. For everyone. At least for everyone who wants it. Stay tuned. 

1.           Barter, P.J., et al., Effects of torcetrapib in patients at high risk for coronary events. New England Journal of Medicine, 2007. 357(21): p. 2109-22.

2.           Kaiser, L., et al., Impact of oseltamivir treatment on influenza-related lower respiratory tract complications and hospitalizations. Archives of Internal Medicine, 2003. 163(14): p. 1667-72.

3.           Doshi, P., T. Jefferson, and C. Del Mar, The imperative to share clinical study reports: recommendations from the tamiflu experience. PLoS Med, 2012. 9(4): p. e1001201.

4.           Achari, V. and A.K. Thakur, Association of major modifiable risk factors among patients with coronary artery disease--a retrospective analysis. Journal of the Association of Physicians of India, 2004. 52: p. 103-8.

5.           Wannamethee, S.G., et al., Metabolic syndrome vs Framingham Risk Score for prediction of coronary heart disease, stroke, and type 2 diabetes mellitus. Arch Intern Med, 2005. 165(22): p. 2644-50.

6.           Ioannidis, J.P.A. and I. Tzoulaki, Minimal and Null Predictive Effects for the Most Popular Blood Biomarkers of Cardiovascular Disease. Circulation Research, 2012. 110(5): p. 658-662.

7.           Paul, S.M., et al., How to improve R&D productivity: the pharmaceutical industry's grand challenge. Nat Rev Drug Discov, 2010. 9(3): p. 203-214.

Research Blogging:
Barter, P., Caulfield, M., Eriksson, M., Grundy, S., Kastelein, J., Komajda, M., Lopez-Sendon, J., Mosca, L., Tardif, J., Waters, D., Shear, C., Revkin, J., Buhr, K., Fisher, M., Tall, A., & Brewer, B. (2007). Effects of Torcetrapib in Patients at High Risk for Coronary Events New England Journal of Medicine, 357 (21), 2109-2122 DOI: 10.1056/NEJMoa0706628
Kaiser, L. (2003). Impact of Oseltamivir Treatment on Influenza-Related Lower Respiratory Tract Complications and Hospitalizations Archives of Internal Medicine, 163 (14), 1667-1672 DOI: 10.1001/archinte.163.14.1667
Doshi P, Jefferson T, & Del Mar C (2012). The imperative to share clinical study reports: recommendations from the tamiflu experience. PLoS medicine, 9 (4) PMID: 22505850
Wannamethee, S. (2005). Metabolic Syndrome vs Framingham Risk Score for Prediction of Coronary Heart Disease, Stroke, and Type 2 Diabetes Mellitus Archives of Internal Medicine, 165 (22), 2644-2650 DOI: 10.1001/archinte.165.22.2644


Ioannidis JP, & Tzoulaki I (2012). Minimal and null predictive effects for the most popular blood biomarkers of cardiovascular disease. Circulation research, 110 (5), 658-62 PMID: 22383708

Paul, S., Mytelka, D., Dunwiddie, C., Persinger, C., Munos, B., Lindborg, S., & Schacht, A. (2010). How to improve R&D productivity: the pharmaceutical industry's grand challenge Nature Reviews Drug Discovery DOI: 10.1038/nrd3078
  PrintPDF