Study: Researchers with Glaxo ties favored Avandia

As mentioned in my previous post on the MIT controversy surrounding an economist's testifying to Congress on healthcare policy without revealing possible economic conflicts of interest that could affect his views, frequently expressed on this blog are concerns about undisclosed conflicts of interest and their corrosive effects upon healthcare (query link).

One major question that arises is the degree to which conflicts of interest can affect the integrity of the scientific literature. Answering this question is more a matter of social science research rather than biomedical inquiry.

One internal medicine resident at the Mayo Clinic took on this challenge and published such a study, a systematic review, in the British Medical Journal. It is entitled "Association between industry affiliation and position on cardiovascular risk with rosiglitazone: cross sectional systematic review" (BMJ 2010;340:c1344).

Mayo Clinical researchers led by the resident, Amy Wang, examined more than 200 articles that appeared after an analysis in the NEJM linked Avandia to a 43 percent increased risk of heart attacks, and a subsequent clinical trial found no greater danger of heart disease. The results are fascinating. From a summary in the Philadelphia Inquirer:

"We aimed to determine whether financial conflicts of interest with pharmaceutical manufacturers could be fueling this fire," wrote the researchers, led by Amy Wang, a resident in internal medicine at the Mayo Clinic in Rochester, Minn. "From our findings, it appears that the answer is yes.

From the BMJ article itself, free full text available as of this writing at link above:

For each article, we sought information about the authors’ financial conflicts of interest in the report itself and elsewhere (that is, in all publications within two years of the original publication and online). Two reviewers blinded to the authors’ financial relationships independently classified each article as presenting a favourable (that is, rosiglitazone does not increase the risk of myocardial infarction), neutral, or unfavourable view on the risk of myocardial infarction with rosiglitazone and on recommendations on the use of the drug.

The "and elsewhere" can be a valuable addition to the capabilities of today's researchers, often enabled via the internet (e.g., non-literature based resources such as conference brochures, personal web pages, corporate and other financial disclosures. etc.) when articles themselves prove unrevealing.

Overall, this is a straightforward methodology, and while potentially subject to bias and expertise issues, nonetheless seems an excellent new contribution to the needed social research on undisclosed COI's.

As to results (see the full article for statistical details I am omitting here):

Of the 202 included articles, 108 (53%) had a conflict of interest statement. Ninety authors (45%) had financial conflicts of interest. Authors who had a favourable view of the risk of myocardial infarction with rosiglitazone were more likely to have financial conflicts of interest with manufacturers of antihyperglycaemic agents in general, and with rosiglitazone manufacturers in particular, than authors who had an unfavourable view. There was likewise a strong association between favourable recommendations on the use of rosiglitazone and financial conflicts of interest. These links persisted when articles rather than authors were used as the unit of analysis, when the analysis was restricted to opinion articles or to articles in which the rosiglitazone controversy was the main focus, and both in articles published before and after the Food and Drug Administration issued a safety warning for rosiglitazone.

While causality is not proven, these results are still stunning. Per the Inquirer's summary:

... Almost 90 percent of scientists who wrote positive articles, reviews, or commentaries about Avandia had financial ties to London-based Glaxo, the study published in the British Medical Journal found.

Almost three of every four authors who expressed negative views of the drug had no financial ties to manufacturers of diabetes medicines, while just 6 percent of those with positive opinions of the drug received no funding or fees from industry.

The authors cite prior literature also demonstrating an associations of COI with pro-industry views (references 2-6; #2 is regarding the mid 1990's calcium channel blocker controversy).

The authors observe that:

In the past decade, research and policy has focused on this association [between COI and pro-industry views - ed.], leading to important progress in policies to manage and encourage disclosure of such financial conflicts of interest.

Their findings about the effects of this "progress" was particularly disappointing, that only about half of the articles carried conflict-of-interest statements:

Of the 202 eligible studies (10 reported original research, 91 were letters, editorials, or commentaries, and 101 were reviews, meta-analyses, or guidelines), 108 articles (53%) included a conflict of interest statement. [The other 47% did not - ed.]

One wonders about the publishers of the articles themselves omitting COI statements, either acknowledging possible COI's or confirming that no COI's exist.

... A total of 90 (45%) of the 202 articles were authored by individuals who had financial conflicts of interest. Of the 90 studies with conflicts of interest, 69 (77%) had a statement disclosing the conflict of interest in the article itself. The other 21 studies with financial conflicts of interest (23%) did not disclose these relationships, which were discovered through searching other publications by the same author or the internet. Three (14%) of these 21 studies published a statement declaring no conflicts of interest. [Oops - ed.]

So, almost one quarter of articles with author COI did not disclose that fact, and several misrepresented the issue.

Finally, the pharmaceutical company's view, also published in the Inquirer article cited above:

A Glaxo spokeswoman said the company posted information and results from all its clinical trials on its Web site.

"It's vital that people have trust in the way we do research and the way it's made public," Jo Revill, Glaxo spokeswoman, said yesterday. "Part of that is sharing data. What we have done is develop policies that will have disclosure and encourage disclosure."

"Encouraging" disclosure of COI's is far too weak a stand for a pharmaceutical company to take on such a critical issue, in my view.

The Mayo researchers wisely conclude:

Disclosure rates for financial conflicts of interest were unexpectedly low, and there was a clear and strong link between the orientation of authors’ expressed views on the rosiglitazone controversy and their financial conflicts of interest with pharmaceutical companies. Although these findings do not necessarily indicate a causal link between the position taken on the cardiac risk of rosiglitazone in patients with diabetes and the authors’ financial conflicts of interest, they underscore the need for further changes in disclosure procedures in order for the scientific record to be trusted.

Finally, I add that these findings are not too surprising. It's human nature not to bite the hand that feeds you. Money is also a powerful stimulant to the brain's rationalization (as opposed to rational) centers.

We as a culture do truly need to understand the potential corrosive effects on the scientific literature of dysfunctions such as ghostwriting, financial and other conflicts of interest, publish-or-perish pressures of academia, and other such social issues.

These practices also simply need to be abolished, for as I have written before on these pages ("Has Ghostwriting Infected The Experts With Tainted Knowledge, Creating Vectors for Further Spread and Mutation of the Scientific Knowledge Base?"), once a "critical mass" of faux knowledge is put into circulation via the scientific literature, even the ostensibly impartial experts' views can become tainted, dependent as they are on that very same scientific literature. Even worse, we don't know what that critical mass is - or whether we've reached it already.

-- SS

Fatty Liver: It's not Just for Grown-ups Anymore

The epidemic of non-alcoholic fatty liver disease (NAFLD) is one of my favorite topics on this blog, due to the liver's role as the body's metabolic "grand central station", as Dr. Philip Wood puts it. The liver plays a critical part in the regulation of sugar, insulin, and lipid levels in the blood. Many of the routine blood tests administered in the doctor's office (blood glucose, cholesterol, etc.) partially reflect liver function.

NAFLD is an excessive accumulation of fat in the liver that impairs its function and can lead to severe liver inflammation (NASH), and in a small percentage of people, liver cancer. An estimated 20-30% of people in industrial nations suffer from NAFLD, a shockingly high prevalence (1).

I previously posted on dietary factors I believe are involved in NAFLD. In rodents, feeding a large amount of sugar or industrial seed oils (corn oil, etc.) promotes NAFLD, whereas fats such as butter and coconut oil do not (2). In human infants, enteric feeding with industrial seed oils causes severe liver damage, whereas the same amount of fat from fish oil doesn't, and can even reverse the damage done by seed oils (3).

So basically, I think sugar and industrial oils are major contributors to NAFLD, and if you look at diet trends in the US over the last 40 years, they're consistent with the idea. Industrial oils are harmful due (at least in part) to their high omega-6 content, which is problematic partially because it disturbs normal omega-3 metabolism. A potential solution to fatty liver is to reduce sugar, replace industrial oils with natural fats, and ensure a regular source of omega-3. I've posted two anecdotes of people rapidly healing their fatty livers using diet changes* (4, 5).

I recently came across a study that examined the diet of Canadian children with NAFLD (6). The children had a high sugar intake, a typical (i.e., high) omega-6 intake, and a low omega-3 intake. The authors claimed that the children also had a high saturated fat intake, but at 10.5% of calories, they were almost eating to the American Heart Association's "Step I" diet recommendations**. Busted! Total fat intake was also low.

High sugar consumption was associated with a larger waist circumference, insulin resistance, lower adiponectin and elevated markers of inflammation. High omega-6 intake was associated with markers of inflammation. Low omega-3 intake was associated with insulin resistance and elevated liver enzymes. Saturated fat intake presumably had no relation to any of these markers, since they didn't mention it in the text.

These children with NAFLD, who were all insulin resistant and mostly obese, had diets high in omega-6, high in sugar, and low in omega-3. This is consistent with the idea that these three factors, which have all been moving in the wrong direction in the last 40 years, contribute to NAFLD.


* Fatty liver was assessed by liver enzymes, admittedly not a perfect test. However, elevated liver enzymes do correlate fairly well with NAFLD.

** Steps I and II were replaced by new diet advice in 2000. The AHA now recommends keeping saturated fat below 7% of calories. Stock up on those skinless chicken breasts! Make sure there isn't any residual fat sticking to the meat, it might kill you. I do have to give the AHA credit however, because their new recommendations focus mostly on eating real food rather than avoiding saturated fat and cholesterol.

Gruber at MIT to Senators: "I Promise To Be Good ... Next Time"

Frequently expressed on this blog are concerns about conflicts of interest and their corrosive effects upon healthcare (query link). We also have expressed concern about the mental gymnastics used by the privileged or those in positions of power to rationalize these COI's and/or justify keeping these COI's from the public's eye.

In a stunning display of academic arrogance, MIT economics professor Jonathan Gruber, who promoted and defended the administration's health care policies before the U.S. Senate Committee on Finance as well as the Health, Education, Labor & Pensions Committee, while collecting $400,000 from HHS for his services, literally blew off a letter of inquiry from Senators Grassley and Enzi regarding possible conflicts of interest. (See this Google search on the terms "jonathan gruber conflict of interest HHS".)

The professor has been a source of support for what many consider improbable arguments of the impact of current health care reform efforts, while being on the HHS dole.

Sen. Grassley and Sen. Enzi have thus written Dr. Susan Hockfield, President of MIT, to "encourage" the professor to be a little more forthcoming in his replies.

The letter to Hockfield and a copy of the original letter to Prof. Gruber can be downloaded here (PDF).

On Jan. 26, 2010 Gruber was asked some very specific questions by letter (summary below, see full letter at link above for details).

He was asked by the Senators:

• To identify each contract, grant or cooperative agreement he has executed with the federal government, including the White House and HHS and its operating divisions, during the past five years. For each contract, he was asked to respond to a series of detailed questions on: specifics of work performed, federal official serving as contact, start/end dates, amount of compensation, plus a request for copies of the contracts, grants, cooperative agreements, and a copy of the deliverables;
• To identify which of the contracts are "sole source", and provide the reasons the contracts were sole source, what official made that determination, and who is the federal official with whom he negotiated the sole sourcing;
• To answer similar questions about contracts with state and local governments and private entities and individuals he had executed in the past five years;
• To identify all interviews or statements made to the media on healthcare reform since the federally funded work began, and if asked during the interview or if he did so voluntarily, to state if he revealed the HHS relationship; and
  
• To answer similar questions on all speaking engagements.
  

To these detailed inquiries, the response was this (emphases mine):

February 23, 2010

Dear Senators Grassley and Enzi:

This letter responds to your letter to me dated January 26, 2010.

I appreciate your commitment to the principle of transparency in government and can assure you that I never intentionally withheld my two HHS contracts from Congress. Indeed, my contracts were publicly disclosed on the federal government website, www.usaspending.gov.

[Prof. Gruber seems to confuse the passive "publicly disclosed" with active "public disclosure," and apparently misses the point that a possible conflict of interest is an obligation of the principal(s) to disclose when they publish or speak, not an obligation of the audience to discover ex post facto via governmental or other websites - ed.]

As you know, my professional research focuses on public finance and health economics and throughout my career I have worked with Democrats and Republicans alike on the critical issue of health care reform. I served as a consultant to then-Governor Mitt Romney in connection with health care reform efforts in Massachusetts and have also testified before the Congress on these issues.

[That worked out rather poorly, according to Massachusetts Treasurer Tim Cahill - ed.]

I was invited to testify jointly by the Majority and Minority Members of the Senate Finance Committee in the Summer of 2008 on the interaction of health insurance and the tax code and have testified more recently as well. In addition, over the past year in connection with my economic analysis work for HHS, I have worked closely with the staffs of both the Senate Finance and the Senate HELP Committees in evaluating various health care reform proposals and assessing how such proposals might be scored by the Congressional Budget Office.

To the best of my recollection, during the course of my health care reform work with Congress, no Member or staffer ever asked me whether I held any government contracts. Given that the staff saw the time that I was devoting to this effort, in retrospect, I suppose I assumed they understood I was being paid for that work.

[Assumed you were being paid ... to the tune of nearly half a million dollars? - ed.]

Both my congressional testimony and the health care opinion pieces that I have authored reflect my longstanding policy views -- and were in no way influenced by my contracts with HHS, which related to technical economic modeling designed to determine the costs of various health care reform policy options.

[Which raises another question: was that 'modeling' influenced by the HHS money? - ed.]

It certainly did not occur to me that anyone would view my HHS contracts as relevant to my independent, long-held opinions regarding the economics of health care reform. [Really? - ed.] I regard the contract work I performed for HHS and my public speaking and writing as being independent and unrelated. But I do recognize the public interest in these disclosure issues and will certainly consider them in the context of future appearances on the Hill or in the media.

[I note that "I will consider them" (i.e., disclosure 'issues') is an ambiguous statement, as opposed to "I will disclose my possible conflicts" - ed.]

Thank you for your work on these issues.

Sincerely,

Jonathan Gruber

In other words, I guess the public does have a right to know about my possible conflicts of interest when a major portion of the US economy's about to be up-ended, and I'll be a good boy and team player - next time. By the way, regarding your detailed questions, you do good work, but get lost.

Academic contempt for disclosure of conflicts of interest seems to know few bounds. I doubt the letter to the MIT President will accomplish very much.

Let's hope the responses from the healthcare IT vendors to Sen. Grassley's inquiries are more forthcoming than this academic one.

Finally, it occurs to me that profligate spending of almost a half million dollars for the sole-source consultation of a single (already salaried) professor raises a concern about cherry-picking for theories supporting a certain point of view, and shows questionable governmental fiscal responsibility.

That is the size of a research grant that could support, say, a significant fraction of the time of a half dozen academic collaborators with a diverse set of theories and views.

-- SS

Disclosure: I have not received a penny from HHS, government, academia, industry, or even Tinkerbell or the Tooth Fairy for these views.

Book Review: The Primal Blueprint

Mark Sisson has been a central figure in the evolutionary health community since he began his weblog Mark's Daily Apple in 2006. He and his staff have been posting daily on his blog ever since. He has also written several other books, edited the Optimum Health newsletter, competed as a high-level endurance athlete, and served on the International Triathlon Union as the anti-doping chairman, all of which you can read about on his biography page. Mark is a practice-what-you-preach kind of guy, and if physical appearance means anything, he's on to something.

In 2009, Mark published his long-awaited book The Primal Blueprint. He self-published the book, which has advantages and disadvantages. The big advantage is that you aren't subject to the sometimes onerous demands of publishers, who attempt to maximize sales at Barnes and Noble. The front cover sports a simple picture of Mark, rather than a sunbaked swimsuit model, and the back cover offers no ridiculous claims of instant beauty and fat loss.

The drawback of self-publishing is it's more difficult to break into a wider market. That's why Mark has asked me to publish my review of his book today. He's trying to push it up in the Amazon.com rankings so that it gets a broader exposure. If you've been thinking about buying Mark's book, now is a good time to do it. If you order it from Amazon.com on March 17th, Mark is offering to sweeten the deal with some freebies on his site Mark's Daily Apple. Full disclosure: I'm not getting anything out of this, I'm simply mentioning it because I was reviewing Mark's book anyway and I thought some readers might enjoy it.

The Primal Blueprint is not a weight loss or diet book, it's a lifestyle program with an evolutionary slant. Mark uses the example of historical and contemporary hunter-gatherers as a model, and attempts to apply those lessons to life in the 21st century. He does it in a way that's empowering accessible to nearly everyone. To illustrate his points, he uses the example of an archetypal hunter-gatherer called Grok, and his 21st century mirror image, the Korg family.

The diet section will be familiar to anyone who has read about "paleolithic"-type diets. He advocates eating meats including organs, seafood, eggs, nuts, abundant vegetables, and fruit. He also suggests avoiding grains, legumes, dairy (although he's not very militant about this one), processed food in general, and reducing carbohydrate to less than 150 grams per day. I like his diet suggestions because they focus on real food. Mark is not a drill sergeant. He tries to create a plan that will be sustainable in the long run, by staying positive and allowing for cheats.

We part ways on the issue of carbohydrate. He suggests that eating more than 150 grams of carbohydrate per day leads to fat gain and disease, whereas I feel that position is untenable in light of what we know of non-industrial cultures (including some relatively high-carbohydrate hunter-gatherers). Although carbohydrate restriction (or at least wheat and sugar restriction) does have its place in treating obesity and metabolic dysfunction in modern populations, ultimately I don't think it's necessary for the prevention of those same problems, and it can even be counterproductive in some cases. Mark does acknowledge that refined carbohydrates are the main culprits.

The book's diet section also recommends nutritional supplements, including a multivitamin/mineral, antioxidant supplement, probiotics, protein powder and fish oil. I'm not a big proponent of supplementation. I'm also a bit of a hypocrite because I do take small doses of fish oil (when I haven't had seafood recently), and vitamin D in wintertime. But I can't get behind protein powders and antioxidant supplements.

Mark's suggestions for exercise, sun exposure, sleep and stress management make good sense to me. In a nutshell: do all three, but keep the exercise varied and don't overdo it. As a former high-level endurance athlete, he has a lot of credibility here. He puts everything in a format that's practical, accessible and empowering.

I think The Primal Blueprint is a useful book for a person who wants to maintain or improve her health. Although we disagree on the issue of carbohydrate, the diet and lifestyle advice is solid and will definitely be a vast improvement over what the average person is doing. The Primal Blueprint is not an academic book, nor does it attempt to be. It doesn't contain many references (although it does contain some), and it won't satisfy someone looking for an in-depth discussion of the scientific literature. However, it's perfect for someone who's getting started and needs guidance, or who simply wants a more comprehensive source than reading blog snippets. It would make a great gift for that family member or friend who's been asking how you stay in such good shape.

Interview on Bizymoms

I recently did a written interview for the website Bizymoms.com. It was the first time I had been invited to do an interview, so I figured what the heck. They bravely posted the interview, despite the fact that my responses could be seen as controversial. You can find it here.