Mass Torts: State of the Art : Mass Torts Lawyer & Attorney : Vorys, Sater, Seymour & Pease Law Firm : Toxic Torts & Wrongful Death

Overcoming Bias and Barking Up The Wrong Tree are both commenting about the recent article "Insightful or Wishful: Lawyers' Ability to Predict Case Outcomes". Four hundred eighty one lawyers were queried about their expectations for one of their cases set for trial. After all was said and done the actual outcomes were compared to the trial lawyers' ex ante predictions and they didn't too to well; not even the ones who'd been out for years. So what gives?

Do lawyers overpromise in order to attract business, as Robin Hanson suspects? I'm sure that's sometimes the case. I once had a front row seat, as it were, that afforded me the opportunity to witness (and my client, formerly the target defendant, to benefit from) another firm's Senior Trial Partner's  first jury trial (and believe me his client had to push him to the courthouse kicking and screaming). He was played like a fiddle by the plaintiffs' attorney and by the time he read his closing from across the room he looked like a cornered animal. The verdict was huge.

Yet I don't think overpromising by underperforming trial lawyers accounts for most of the poor predictions. Rather, they likely fell prey to all too common faulty heuristics. They've spent weeks building a narrative that fits the facts and the law and the story ends happily ever after for their clients. And the bad guys' story? They've spent as much time or more poking holes in it so that for every doubt imagined a counterpoint leaps effortlessly from memory. Why isn't this just the usual sort of programmed reasoning exercise that makes most people poor prognosticators?

Finally, how are cases resolved? If settled isn't it usually on terms that nobody's especially thrilled about? And if they're not settled the distribution of outcomes, at least in my experience, tends to be U-shaped with a bunch of $0 verdicts on the left and a bunch of "ring the bell" verdicts on the right and not too many "kiss your sister" verdicts in between. I had a young lawyer recently tell me what he estimated the expected value of his case ought to be. The problem, of course, is that once the bell rings twice (or whatever jurors do in your jurisdiction to signal they've got a verdict) you're about to be, like the owner of Schrodinger's cat, either really really happy or really really sad. When the possible is collapsing into the certain why should it be a surprise that lawyers usually get it wrong when the odds of getting it right are against them?

• Email This
• Print

The incidence of food allergies in children is rising. Wheat, milk, egg, fish, peanut, walnut, shellfish and soy allergies have led to recalls of pork, turkey, cream of wheat mushroom soup, roast beef, ice cream and corn pasta in recent months. What's behind the increase in allergies?

There are at least two good hypotheses for which there's sound evidence. First, despite what our pediatrician told us, it's probably a good idea to introduce babies to e.g. cow milk sooner rather than later (see "Early Exposure to Cow's Milk Protein is Protective Against IgE-Mediated Cow's Milk Protein Allergy") and make sure they get a large enough dose to produce tolerance as it's apparently the low doses of say peanuts that lead to sensitization and allergy (see e.g. "Peanut Sensitization and Allergy: Influence of Early Life Exposure to Peanuts").

The second emerging hypothesis, another of the increasingly common "Grandma was right" sort of ideas, is that lack of sunshine is also responsible for the rise in food allergies in children. It turns out that vitamin D is crucial to a properly functioning immune system and without it you wind up with a gut full of the wrong sorts of bacteria behaving badly. (See "Potential Mechanisms for the Hypothesized Link Between Sunshine, Vitamin D, and Food Allergy in Children" and "The Role of the Gut Mucosal Immunity in the Development of Tolerance Versus Development of Allergy to Food").

And while we're on the topic of the consequences of this needless epidemic of vitamin D deficiency in the U.S. due to four decades of anti-sun/anti-reason activism you should also read: "North-South Differences in U.S. Emergency Department Visits for Acute Allergic Reactions", "Are Active Sun Exposure Habits Related to Lowering Risk of Type 2 Diabetes Mellitus in Women, a Prospective Cohort Study?" and "Vitamin D and Risk of Cognitive Decline in Elderly Persons" along with "Vitamin D: A Place in the Sun?" and "Vitamin D in Asthma: Panacea or True Promise?"

The takeaway here is that by overreacting to rare and likely uncontrollable risks we've been stampeded right into far more common and otherwise avoidable risks. So who should be liable to all of the eggshell plaintiffs manufactured out of junk science? Should it be the companies whose products would not have caused harm but for the activists or should the activists be called to account for what they have done?

• Email This
• Print
• Trackbacks

Though riddled with clever falsehoods and logical fallacies, "The Story of Cosmetics" is brilliant rhetoric - which is to say it's extremely persuasive. The narrator comes across as everywoman - down to earth, smart, earnest, caring and fair yet resolute about protecting families; and, her conclusions follow obviously from her premises.The graphics are especially well done - they enhance what the narrator is saying and add humor and wit. Finally, and importantly, it's short. Before my first trial my Dad told me to "say what you need to say or ask what you need to ask and then sit down and shut up; too many lawyers babble on trying to prove how smart they think they are and accomplish nothing besides annoying the jury and obscuring whatever point they're trying to make." Two decades on I can confirm all aspects of that advice.

Deconstructing this work of rhetorical art needs to be done piecemeal (if only because I have a present ion to complete and thereafter a plane to catch) so let's start at the beginning (well, not the beginning beginning - that's just the anti-consumerism ideological waypoint - I mean the "are these products toxic" beginning). Anyway, the narrator says she loves her shampoo and one day got to wondering what was in it. The first thing she noticed upon reading the label was that it had chemicals in it; specifically sodium laureth sulfate, tetrasodium EDTA and methylisothiazolinone. So she took the list to some unidentified "scientists" who advised that her shampoo "contains a chemical linked to cancer!" Let's look at this claim.

As luck would have it the International Journal of Toxicology just published "Final Report of the Amended Safety Assessment of Sodium Laureth Sulfate and Related Salts of Sulfated Ethoxylated Alcohols". Not surprisingly it's a soap that may irritate your eyes but if properly formulated won't even do that. It's certainly not a carcinogen which is what the American Cancer Society, OSHA and EPA have been saying for years to anyone willing to ask.

And what about tetrasodium EDTA? Before I answer that one I'd bet it's probably not a coincidence that the first two chemicals mentioned have sodium in the name since sodium, as in sodium chloride, has become a politically incorrect element just as California's state rock has become a politically incorrect mineral. In any event there's no evidence that EDTA is a carcinogen. On the other hand, when your dentist uses it to inhibit biofilm formation in your mouth it's actually an anti-carcinogen. Then there's the fact that EDTA is used to remove toxic metals from the body. Hmmm.

Finally, there's methylisothiazolinone, a biocide that keeps fungi and bacteria from growing in your shampoo. A quick trip through the ether to the U.S. National Library of Medicine leads to the brand new "Final Report of the Safety Assessment of Methylisothiazolinone". Reaffirming previous findings the report concludes that in the doses used in personal care products methylisothiazolinone is safe. But there's something very important here that you'll miss if you're not paying attention to the big picture; and that something is the weight given in vivo versus in vitro data.

How should chemicals be tested for toxicity? Typically it's been done on animals but because of political opposition to animal testing and a 20th century belief in genetic determinism a big yet largely unnoticed push has been underway to have EPA, NTP, etc shift to doing toxicity testing on human cell lines in vitro. So far such cell line in vitro testing has repeatedly managed to find signs of cellular damage from a host of substances never before associated with toxicity. Methylisothiazolinone is one such example.

Despite the fact that neurotoxicity has never been associated with methylisothiazolinone in humans, rats, mice, etc in vitro testing found that it produced damage to a line of human neural cells. What gives?

What gives, of course, is that just as a bacterium responds to its environment differently depending on whether it's alone or in a biofilm so it's hardly a surprise that a single cell would respond to stimuli differently than would the entire organism of which it is but a vanishingly small part. Thus the decision that methylisothiazolinone is safe thanks to the weight of the in vivo data. Therefore, as we shift away from the tried and true methods of toxicity testing and towards a new model which, in light of recent discoveries regarding epigenetics and the microbiota which run a good portion of our lives, be prepared for many many more baseless health scares; and the worst part about it is that without the ability to test the hypotheses generated on actual animal models refuting even the most absurd claims will only get harder.

• Email This
• Print
• Trackbacks

The Institute of Medicine's "Ethical Issues in Studying the Safety of Approved Drugs: A Letter Report (2010)" is a slog punctuated by the occasional eye-roller. Perhaps because of the presence of undefined ethical issues, technical discussions of drug trials are interrupted with fawning nods to stakeholders "who have kinds of knowledge different from those of technical experts" and a promise to incorporate their different ways of knowing. Sheesh. Anyway, the report does suggest something profoundly important and that is a shift away from the randomized controlled trial (the so-called gold standard) to observational epidemiological studies when making decisions about whether an approved drug ought to be pulled from the market.

Observational epidemiology has been on life support for years. It worked very well on the big risks but when it was turned loose on smaller risks, or worse yet, simply unleashed upon mounds of raw data without any a priori hypothesis to be tested, it failed altogether and often as not wound up falsely indicting random aspects of daily life as wanton spree killers. The failures of course were due neither to the math nor the principles but rather to the misuses to which it was put - especially by expert witnesses who knew better. Nevertheless, finding a role for epidemiology in a world of small risks has been a topic of much debate. Online, and free, you can find the debate in full at "The Triumph of the Null Hypothesis: Epidemiology in an Age of Change" and Jan P Vandenbroucke's commentary, "Seen from Another Angle".

For the Institute of Medicine (IOM), at least given qualms about the ethics of randomized controlled trials particularly after questions have been raised about a drug, Vandenbroucke's view has apparently prevailed. The IOM concluded that "under specific circumstances, observational studies may be adequate to ... identify the presence of an important safety issue ..."; and further that such studies "could provide useful and valid estimates of the risk associated with a safety signal." And, when a drug trial is ongoing, e.g. TIDE, "if new evidence from any source, including the trial itself, is determined to be sufficiently compelling to ground a policy decision without waiting for additional new information, allowing the trial to continue would be unethical."

So, if observational epidemiology prompted by a potential signal of a serious risk were to suggest a small risk, the drug could be pulled from the market and the only test capable of answering the question of whether or not it actually posed a risk would be stopped and prohibited from being run? It sounds like it. If that's the case observational epidemiology has come roaring back to attain a position of power and prestige not held in many years.

• Email This
• Print

Running sophisticated statistics software over a pile of data drawn from the opinions of an appellate court does not guarantee a sophisticated analysis of the court's decision-making. In fact, given what I've read in a couple of fairly recent law review articles, about the only thing it guarantees are unwarranted conclusions and no analysis whatsoever.

I had high hopes for the testing of hypotheses about legal reasoning. Take for example the work of Jeffrey Rachlinski. He was able to demonstrate that judges have some of the same cognitive blind spots as the rest of us; the most notable of which being hindsight bias. (See e.g. "Inside The Judicial Mind"). But this new fad of indiscriminately dredging for correlations and publishing uncritically the inferences generated thereby is very troublesome.

Let's take a suggestion from one law review article that a certain court is biased because the win/loss ratio for plaintiffs and defendants deviates significantly from what would be expected of an unbiased court. Specifically, that the defense prevailed in more than 68% of the cases tallied implying to the author that something significant was afoot since the observed distribution was more than one standard deviation beyond the expected. Upon what premise must such a conclusion rest? Either that justice is or ought to be decided by casting lots (in other words that the distribution of wins and losses should look like the results of flipping a coin a large number of times and so appear as a Gaussian distribution) or that cases appear randomly before the court. The former denies the existence of justice as we know it and the latter denies the facts - plaintiffs often press novel claims, seek to overturn or plead around statutes (like tort reform) or try to hold on to huge verdicts - all of which might reasonably be expected to influence the likelihood of success on appeal. Nevertheless, none of these assumptions or possible biases and confounders are discussed anywhere in the paper.

Another recent article demonstrates the misuse and abuse of statistics. In it the author claimed to have uncovered a power law behind certain judicial decision-making. Finding a hidden power law is considered very sexy these days. They're different than merely normal distributions; they explain an increasing number of natural phenomena (e.g. most prominently some economic  conditions in the current troubles - see: Pareto distribution); and, best of all, they allow you to torture your data with even more impressive-sounding statistical tools.

The problem however is that putting fancy tools in the hands of someone who doesn't know how to use them cannot lead to anything worthwhile. And so it was with the second paper that had me despairing. On a log-log plot a power law will reveal its distribution as a straight line. But while a straight line on a log-log plot may be necessary for the presence of a power law it's not sufficient; yet that's stated no where in the paper. But that's a minor quibble compared to the real problem with the analysis. Rather than plotting the data on a scatter graph and inferring that a power law might lie beneath, the author apparently put the data on a log-log plot, forgot it was a log-log plot and ran a linear regression on it thereby forcing a nice straight line where there was none - the result being a sort of tautology-by-statistics. 

Don't get me wrong. The discovery that certain behaviors could be explained and predicted mathematically would be a big deal. But when you see a vigorous debate in the scientific literature about whether bacterial foraging behavior (swim straight vs. tumble and then swim somewhere else) should be described by a power law curve rather than Gaussian bell curve it makes you wonder whether something as complicated as the administration of justice can ever be reduced to a simple formula. Human scale biological systems are fantastically complex and I'd expect that the approach a judge takes to her task of doing justice, the application of principles - our way of dealing with complexity and uncertainty, would be even moreso.

• Email This
• Print
• Trackbacks

Yuck! Tastes like CBPR.

CBPR? I hadn't heard of it either; not until this week anyway. It's "community-based participatory research" and the idea is to get activists and scientists together to do research into things like the cause of higher than average breast cancer rates in some high net worth communities. The activists want the scientists to prove that pesticides are causing breast cancer in their communities; and so the scientists promptly set out to falsify the activists' claims. Why wouldn't it turn out well?

The answer can be found in "A Review of Advocate/Scientist Collaboration in Federal Environmental Breast Cancer Research". It turns out that "effective" CBPR requires a different sort of "inquiry paradigm." You see "[t]he positivist paradigm remains dominant in much scientific research, emphasizing objective knowledge that is separate from the knower and can only be uncovered through a scientific method of inquiry that is neutral and bias-free." "CBPR challenges this paradigm by contextualizing scientific research within particular communities, including and legitimizing advocates' knowledge, understandings, and priorities regarding issues by which they are personally affected." (Ibid at pg 17).

So researchers, schooled in the long history of how biases, prejudices and failures to challenge closely held beliefs have thwarted science and medicine in the past, are to drop everything they hold dear? Or is it that advocates are to drop their beliefs, acquiesce to all the money they lobbied for being spent on an effort to falsify those beliefs; and, after seeing them falsified, to say "Nevermind", take the hit to their reputations and set about constructing a new narrative for their lives? The former rejects the scientific method; the latter, human nature. 

• Email This
• Print

The American Cancer Society is calling for new research to settle the issue of whether or not twenty different agents do indeed cause the types of cancer in which they've been implicated. The twenty are:

(1) Lead and lead compounds; (2) indium phosphide (used in many flat screen TVs); (3) cobalt with tungsten carbide; titanium dioxide; (4) welding fumes; (5) refractory ceramic fibers; (6) diesel exhaust; (7) carbon black; (8) styrene oxide and styrene; (9) propylene oxide; (10) formaldehyde (does it cause leukemia?); (11) acetaldehyde; (12) formaldehyde; (13) methylene chloride; (14) trichloroethylene; (15) tetrachloroethylene; (16) chloroform; (17) PCBs; (18) DEHP (a phthalate); (19) atrazine (a herbicide and the subject of a coordinated attack by various activists groups resulting in a new EPA review); and, (20) shift work (the presumed exposure being "light at night" leading to a disruption of circadian rhythms and the most commonly associated malignancy being breast cancer).

You can find the press release here: Report Outlines Knowledge Gaps for 20 Suspected Carcinogens; and you can find the IARC report summarizing past rationale for assigning these suspected carcinogens to groups 2A - 3, the new evidence forming the basis for the recommendation that the status be updated and the sorts of epidemiological and mechanistic studies necessary to answer the question of whether they ought to be added to the list of 107 Group 1 agents known to be carcinogenic to humans, here: Identification of Research Needs to Resolve the Carcinogenicity of High-Priority IARC Carcinogens.

• Email This
• Print
• Trackbacks

Sometimes The New York Times just publishes the press releases of zillionaire trial lawyers and fellow traveler advocacy groups in its Science and Health sections. Most of the time though, when power and money aren't apparently in play anyway, The Gray Lady does the best job in the business of reporting on scientific advances.

One advance that we've been covering for the last year is the staggering realization that most of the cells in "our" body aren't ours; that much of the heavy lifting that keeps us healthy is done by the aliens inside of us; and, that genetic determinism is a fairy tale.

Read "How Microbes Defend and Define Us" for an excellent introduction to the topic of the Mighty Microbe that saves the day.

• Email This
• Print

Despite a steady drip drip drip of out-of-context memos, disconnected snippets of depositions and dire predictions that rose to a torrent last night, the FDA's panel of experts today voted 20 - 13 to keep Avandia on the market, though, most suggested, with more warnings. More significantly, despite vicious ad hominem attacks on respected academics and physicians, the panel also voted to complete the TIDE study - one of those "gold standard" studies that should in 2015 produce at last some definitive data on the questions of whether Avandia poses an unacceptable risk for diabetics and whether it produces a better outcome than a competitor.

Of all the objectives of those who lead such attacks their effort to declare science settled and to outlaw further experiments that might falsify their claims is the most frightening. On the other hand, when you consider what happened with the breast implant litigation you have to understand that by waiting for the truth those who fed The New York Times stray documents and quotes from discovery would only put their businesses at risk. Fama, malum qua non aliud velocius alium.

• Email This
• Print

Well, that didn't take long. A new study published in Lancet Oncology titled "Angiotensis-Receptor Blockade and Risk of Cancer: Meta-Analysis of Randomized Controlled Trials" is the result of yet another ex post assessment of mountains of data from trials of pharmaceuticals. This one found a piddling 1.08 risk ratio based on a 95% CI of 1.01 to 1.15. Run for your lives sort of stuff this ain't.

But of course that didn't stop some from exploiting the study; immediately. In something called "People's Pharmacy" the Houston Chronicle on-line led with the following story on the front page: "Cancer Link Feared With Blood-Pressure Medicine." In response to a letter allegedly from a stressed out reader the People's Pharmacy uncritically notes that the paper found "a modestly increased risk of new cancer" and helpfully suggests 8.5 ounces of beetroot juice as an alternative remedy for high blood pressure.

Presumably they're referring to "Inorganic Nitrate Supplementation Lowers Blood Pressure in Humans. Role for Nitrite-Derived NO". Nowhere in the paper do the authors suggest that patients abandon their blood pressure medications in favor of an unknown number of cans of "beetroot juice". Yet that's almost certainly what many will do.

After decades of effort heart disease, which felled my grandfather at 49, is at last receding as a cause of premature death in Americans. It would be tragic indeed if the product of a data-mining expedition or two reversed the habits of those who have benefited so much from that effort.

• Email This
• Print

"'What we have heard back from drug companies is that, if it has the word 'PPAR' in it, it's dead,” says Bruce Spiegelman, a cell biologist at the Dana-Farber Cancer Institute." That's from "Diabetes Drug Woes Spell Trouble for the Entire Drug Family." The most famous PPAR-gamma drug out there is, of course, Avandia.

Why hope PPAR-gamma drugs aren't dead? Well, read "in Vitro and In Vivo Therapeutic Efficacy of the PPAR-gamma Agonist Troglitazone in Combination With Cisplatin Against Malignant Pleural Mesothelioma Cell Growth". The PPAR-gamma agonist not only inhibited malignant pleural mesothelioma cell growth it also lengthened survival.

• Email This
• Print

Today The New York Times, which dutifully fanned the flames of the 2007 "prescription drug crisis" started by those pushing for greater FDA powers and fewer new drugs, published "Caustic Government Report Deals Blow to Diabetes Drug". In essence it reports Dr. Thomas Marciniak's criticism of the RECORD study which in 2007 led the FDA, despite congressional histrionics, to vote 22 - 1 to keep Avandia on the market. What the NYTimes is talking about is this.

What sets off the alarms, to a mass tort lawyer anyway, is slide 22. Why, asks the good doctor, should you believe his numbers? After all he declares that he has "nothing to hide" and that "[n]either my job nor (for me) $100,000,000's are riding on the results." The other slides evidence an effort to dig, but not too much, into the data and upon finding seeming errors to imply, without saying so, that the manufacturer somehow managed to beguile honest researchers from around the world into signing off on bad science. It's the kind of drama you'd expect to see from a certain sort of expert witness testifying at the courthouse in Jefferson County; but hardly the sort of presentation typical of scientific gatherings. Then again the FDA has been hyper-politicized so maybe this is the new normal.

Anyway, the implication that this is the "Government Report" is highly misleading. It is in fact but one of many government reports (if by that we agree to mean presentations generated by government employees/contractors). Indeed, another "Government Report" addresses Dr. Marciniak's claims. You'll find it here.

Dr. Marciniak did the easy thing. Post hoc he rummaged around for evidence of errors that would undermine the RECORD study. When examining the outcomes of thousands of people based on many times that number of documents he found a few seeming inconsistencies. Anyone who does mass tort litigation knows that if a bad data point or two were enough to refute any study then we wouldn't have much to talk about down at the courthouse.

What's also interesting is the fact that throwing in the extra assumed heart attack episodes (even the ones for which there were no biomarkers confirming same) the study still only shows a small increase (1.38) that is statistically significant by the barest margin (C.I. 0.99 - 1.93) (i.e. "not") and still it does not support the Nissen hypothesis (to say nothing of the study of 227,000 Medicare patients that rejects it).

But most interesting of all is the data on the only question we really, ultimately, want answered. Do the people taking the medication live longer, or die sooner, than those who don't? By that measure, the protesting doctor's numbers still show that patients on Avandia were 14% less likely to die than those who weren't. By this most critical measure, even considering the data cherry picking of Dr. Marciniak, the results for Avandia are "reassuring" - according to that other government report.

• Email This
• Print

Part of what Congress passed in response to the perceived prescription drug crisis of 2007, Title IX, Section 921 of the Food and Drug Administration Amendments Act 2007 (FDAAA) (121 Stat. 962), includes a directive that the FDA "conduct regular, bi-weekly screening of the Adverse Event Reporting System [AERS] database and post a quarterly report on the Adverse Event Reporting System Web site of any new safety information or potential signal of a serious risk identified by Adverse Event Reporting System within the last quarter."  A single sufficiently serious adverse event may be enough to constitute a "potential signal".  The number of potential signals being detected are increasing rapidly and on the most recent quarterly Potential Signals report thirteen additional drugs have been listed including such widely prescribed medications as Zithromax (the highly effective Z-pack) and Premarin for producing signals of serious risks of liver failure and angioedema respectively.

What will be the impact of releasing what is essentially raw data divorced from any analysis of its meaning or discussion of the plausibility of what it implies?  In other words, what good will come from this sort of transparency?  Other than tempting the unwary to the wrong conclusion thanks to the post hoc ergo proptor hoc logical fallacy it's hard to see any good coming of this; unless of course you're a personal injury lawyer.  Google "Zithromax lawyer" and you'll find that though the new potential signal report is only a few days old there are already numerous "Zithromax attorneys" anxious to represent you in your claim for liver failure.

At the end of the day it may well be that the so called prescription drug crisis that the Congress was responding to in 2007 was in fact merely a prophecy only now being fulfilled thanks to the AERS Potential Signals reporting system.

• Email This
• Print
• Trackbacks

The American Cancer Society has just published "Cancer Statistics, 2010", its annual estimates of the number of new cancer cases and deaths expected in the United States. Did you know that since 1990 the rate of death from all cancers combined for men of all races combined has fallen 21%? Yet thanks to a longer life expectancy and advances in preventing people from dying from e.g. heart attacks 44 of every 100 men will develop an invasive cancer in their lifetime. Thanks to smoking women are catching up to men on the categories you don't want to be in and falling behind in the ones you do, like decreasing death rates.

After accidents comes cancer on the list of causes of death in children. The good news is that the overall 5-year survival rate is now 81% and the trend is in the right direction. For acute lymphocytic leukemia (ALL) and non-Hodgkin lymphoma (NHL) the 5-year survival rates are now 89% and 95% respectively.

In all, 562,875 Americans died of cancer in the most recent year for which all data has been collected. For men and women aged 40 - 79 cancer is leading cause of death. In fact, it's the leading cause of death among everyone under the age of 85 combined beating out heart disease by a body count of 475,211 to 380,791.

There's lots more in the report and it's free at the link above.

• Email This
• Print

Three years ago Dr. Steven Nissen published "Effect of Rosiglitazone on the Risk of Myocardial Infarction and Death from Cardiovascular Causes" in which he reported that those taking Rosiglitazone (Avandia) had about a 40% increased risk of acute myocardial infarction (heart attack). The study was a meta-analysis of published and unpublished data and drew extensively from the manufacturer's unpublished data.

The media simultaneously reported the findings and propagated the following narrative: The pharmaceutical companies collect vast quantities of data; publish only what supports their products; and are at best willfully ignorant of the risks of their products which risks are right before their eyes if only they would look. Dr. Nissen who "has a statistician’s zeal for drilling deep into clinical data, seeking signs that some widely used drugs pose undisclosed risks to patients" was made the hero of the drama. Congress got involved by beating the drums of safety and transparency and demanding that the FDA pay more attention to ensuring that pharmaceuticals are safe.

The FDA reassessed Avandia and panel members voted 21-3 to keep it on the market.

In February of this year, a Congressional investigation conferred near-martyr status on Dr. Nissen after it was revealed that he had secretly tape recorded a meeting with the representatives of the maker of Avandia. The same month, in European Heart Journal, he published the editorial "The Rise and Fall of Rosiglitazone". Meanwhile everyone was waiting for the results of a huge study of Avandia.

Last week that study, "Risk of Acute Myocardial Infarction, Stroke, Heart Failure, and Death in Elderly Medicare Patients Treated with Rosiglitazone or Pioglitazone" (interestingly, for several days JAMA made you click through an ad for a competitor's type 2 diabetes product to get to the article) was published. And what was the risk of heart attack among those elderly patients on Avandia? Essentially no different than those on the older medication - a statistically insignificant 1.06 increase - especially so given the fact that almost three quarters of all type 2 diabetics wind up dying of heart disease in any event. So, was Avandia cleared? Nope. In fact, the calls for removing it from the market have grown even louder.

The new study, while rejecting the original claim that Avandia causes heart attacks, raised the hypotheses that Avandia causes congestive heart failure and stroke and increases the rate of mortality overall. Yet even though the numbers of patients involved in the study (their records were simply culled from Medicare databases) was large (227,000) the increases were relatively small and by the time the increases appeared (after six to fifteen months of medication) only a small and rapidly decreasing fraction of the original cohort were actually still on one of the two medications.

At almost the same time Dr. Nissen published an updated meta-analysis ("Rosiglitazone Revisited: An Updated Meta-Analysis of Risk for Myocardial Infarction and Cardiovascular Mortality"  also free and also, though published in the Archives of Internal Medicine, only after an ad for the same competitor's type 2 diabetes product) that purports to show that Avandia increases the risk of heart attack but doesn't increase the risk of mortality. What? Hey, wait a second! How can ... So, siding with those advocating "safety over certainty" the New York Times quickly editorialized in favor of those demanding Avandia be removed from the market.

The LATimes went so far as to publish a piece calling for "an immediate moratorium on sales as soon as a credible study raises questions about safety." In that same piece the journalist was posed the following question by a researcher being interviewed: "Suppose a drug saves five people and kills one person. Do you keep it on the market?" His answer was "I know this: If that one person killed is my loved one - or yours - the answer is readily apparent."

Yikes. Really? If a data dredge shows any risk of a fatal outcome then the drug should be pulled from the market no matter how many people are killed in the process? We'll save for another day the question of how thinking about risk goes so badly astray. For now though consider how likely it is (yes likely - and in fact, if the number of endpoints examined are sufficiently large, how almost certain it is) that a risk will be found where none exists. Start with "Data Dredging, Bias, or Confounding: They Can All Get You Into the BMJ and the Friday Papers" and then for more on the perils of statistical deep drilling read "Your Intuitions Are Not Magic" and the links therein.

At the end of the day the issue isn't safety versus certainty. Claiming that pharmaceutical manufacturers are insisting upon certainty before warnings are issued or products are pulled is just a straw man argument. The usefulness of statistical analyses of medical outcomes has not been added to "death and taxes".The real question is which course is less uncertain - making vital judgments on the basis of large randomized controlled trials like the one due out on Avandia in 2015 or on the basis of data dredges? The answer ought to be obvious.

• Email This
• Print