Yet Another Opinion in Which a Court Mistakes Hypothesis for Theory

While some may imagine that scientific hypotheses are the product of highly educated people with brilliant minds drawing straightforward inferences from compelling evidence the fact remains that all scientific hypotheses are nothing more than guesses; and as every middle schooler taught the scientific method knows, even the best pedigreed hypotheses are usually false. On the other hand, sometimes it's the hypothesis with the most dubious provenance that gets promoted to the status of scientific theory (i.e. one that has survived rigorous testing and is powerfully explanatory) as in the case of benzene's structure:

I was sitting writing at my textbook but the work did not progress; my thoughts were elsewhere. I turned my chair to the fire and dozed. Again the atoms were gambolling before my eyes. This time the smaller groups kept modestly in the background. My mental eye, rendered more acute by the repeated visions of the kind, could now distinguish larger structures of manifold confirmation: long rows, sometimes more closely fitted together, all twining and twisting in snake like motion. But look! What was that? One of the snakes had seized hold of its own tail, and the form whirled mockingly before my eyes.

Because a hypothesis is nothing more than the assembly (by hard work or daydreaming) of a few bits of what is known/believed into a plausible narrative that explains some phenomenon (e.g. gastric lymphoma), because so little is known about the causes of a complex disease like gastric lymphoma such that the discovery of H. pylori suddenly and completely overturned prior views about its causes, and because we can't know (or factor into our hypotheses) what we don't know (you've heard of the human gut microbiome but what about the human gut virome?) hypotheses are nothing more than speculation. That's why every epidemiological study you've ever read puts the burden of proof squarely on the hypothesis and resolves all doubt in favor of the "null hypothesis" (i.e. the hypothesized causal agent has no effect).

Unfortunately many courts either don't understand the difference or refuse to distinguish between hypothesis and theory. A recent example is Walker v. Ford. In Walker plaintiff's expert was allowed to opine on the basis of his hypothesis that asbestos is a cause of Hodgkin's lymphoma and thereafter to deduce from another of his hypotheses (Hodgkin's lymphoma is caused by either Epstein-Barr virus, smoking or asbestos) that plaintiff's lymphoma must have been caused by asbestos as he hadn't the virus and didn't smoke. And it isn't just another case of a court conflating hypothesis generation (guessing) with the scientific method (testing guesses) so that guesswork by a properly credentialed witness is turned into a "scientifically valid method" and Rule 702 can be deemed satisfied. It's worse. Not only has the hypothesis that asbestos causes Hodgkin's lymphoma never been verified, it has in fact been repeatedly tested and serially refuted. Furthermore, the most important observation that spawned the hypothesis in the first place (an increased risk of gastric lymphoma among a sample of asbestos workers) has never been reproduced (and will never be reproduced) because when the study was done nobody outside two researchers in Australia even knew H. pylori existed much less to look for it in gastric lymphoma patients - several years would elapse between its discovery and the determination that it is worldwide the leading cause of gastric lymphoma.

The general causation opinion of plaintiff's expert rested on these studies:

1) Cancer Morbidity of Foundry Workers in Korea. A slight increased risk of stomach cancer and non-Hodgkins's lymphoma was found among foundry workers exposed to a laundry list of things including asbestos. No exposure assessment was done for any substance and no increase in Hodgkin's disease was reported. The mortality study of the workforce published this year isn't any more persuasive - here's the SMR table for malignant diseases: SMR table.

2) Extranodal marginal zone lymphoma of mucosa-associated lymphoid tissue type arising in the pleura with pleural fibrous plaques in a lathe worker. Guess what? Asbestos isn't the only cause of pleural plaques and so I stopped reading this article when I got to "He had not been exposed to asbestos."

3) Asbestos exposure and lymphomas of the gastrointestinal tract and oral cavity. This is the study mentioned above that suffers fatally from the understandable ignorance of the confounder H. pylori though it also appears to have the multiple comparison problem as evidenced by the fact that subgroupings of lymphomas, here GI and oral, produced a higher risk than for lymphomas in general. Finally, being a case-control study, there was no estimation of exposure in any of the cases.

4) Does asbestos exposure cause non-Hodgkin's lymphoma or related hematolymphoid cancers? A review of the epidemiologic literature. I didn't get past the abstract which concludes that a review of the literature reveals "no increased risk of NHL (non-Hodgkin's lymphoma) or other HL-CAs (hematolymphoid cancer) associated with asbestos exposure."

Not discussed in Walker but apparently the last nail in the asbestos-causes-lymphoma hypothesis' coffin (and the last sign of any scientific interest in this apparently dead issue) occurred 10 years ago with the publication in the Annals of Epidemiology of Occupational asbestos exposure and the incidence of non-Hodgkin lymphoma of the gastrointestinal tract: an ecologic study. The study found "no support for the hypothesis that occupational asbestos exposure is related to the subsequent incidence of GINHL (gastrointestinal tract non-Hodgkin's lymphoma).

These articles along with the expert's belief that "as long as asbestos reaches an area, regardless of where it is, it can cause different types of cancer" and asbestos can make its way to the lymph nodes, were all he needed to opine that asbestos causes lymphoma including plaintiff's Hodgkin's lymphoma (because after all "a lymphoma is a lymphoma" save "for therapeutic purposes"). That's too much nonsense to unpack in one blog post so I'll just focus on the claim that wherever asbestos goes in the body it causes cancer. The Institute of Medicine was tasked with answering this very question - is there evidence for a causal relationship to asbestos for cancer of everything from the larynx to the rectum - and generally found that what was in the literature was suggestive but insufficient to reasonably conclude that there is a causal link. See: Asbestos: Selected Cancers.

To save plaintiff's expert and his hypothesis the appellate court held that it doesn't matter if an expert's conclusions are correct. All that matters is that the method whereby he reaches his opinion is reliable, and plaintiff's expert's method, guessing about the cause of Hodgkin's lymphoma by creating a narrative about the causation of Hodgkin's lymphoma from a few studies (that didn't actually study Hodgkin's lymphoma) counts as a reliable one. But who, other than the hopelessly ironic, would label as "reliable" a method (i.e. the guessing that constitutes a scientific hypothesis) of causal determination the product of which is usually incorrect? Recall that not only are most scientific hypotheses false but that even most of those with a statistically significant chance of being true are probably false.

Only scientific theories get the Seal of Reliability, which is to say they make predictions on which you can rely. And they gain that status only by being put to the test, and passing; and by passing I mean that the predictions they make actually come to pass. So what prediction would follow from "asbestos exposure causes Hodgkin's disease"? Wouldn't it be "people exposed to asbestos are more likely to get Hodgkin's disease than those who aren't"? And what follows from the fact that no study of asbestos-exposed workers has shown an increased risk of Hodgkin's lymphoma? That the claim "asbestos causes Hodgkin's disease" isn't reliable.

So if hypotheses are unreliable in general because by definition they have not been tested, and if the specific hypothesis "asbestos causes Hodgkin's lymphoma" is unreliable because it has been tested and failed to predict the future it entails, in what sense is the opinion of Walker's expert "reliable"? Let me know if you figure it out.

Need Something To Attack The Reliability Of Animal Studies?

The Reference Manual on Scientific Evidence, Third Edition has the following to say about animal studies:

Animal studies have a number of advantages. They can be conducted as true experiments, and researchers control all aspects of the animals' lives. Thus, they can avoid the problem of confounding, which epidemiology often confronts. Exposure can be carefully controlled and measured. Refusals to participate in a study are not an issue, and loss to followup very often is minimal. Ethical limitations are diminished, and animals can be sacrificed and their tissues examined, which may improve the accuracy of disease assessment. Animal studies often proved useful information about pathological mechanisms and play a complementary role to epidemiology by assisting researchers in framing hypotheses and in developing study designs for epidemiologic studies.

Animal studies have two significant disadvantages, however. First, animal study results must be extrapolated to another species - human beings - and differences in absorption, metabolism, and other factors may result in interspecies variation in responses. For example, one powerful human teratogen, thalidomide, does not cause birth defects in most rodent species. Similarly, some known teratogens in animals are not believed to be human teratogens. In general, it is often difficult to confirm that an agent known to be toxic in animals is safe for human beings. The second difficulty with inferring human causation from animal studies is that the high doses customarily used in animal studies require consideration of the dose-response relationship and whether a threshold no-effect dose exists. Those matters are almost always fraught with considerable, and currently unavoidable, uncertainty.

It turns out that there's a third significant disadvantage and one that ranks ahead of the other two. It's nicely summed up in the title of a new article in Nature - "Animal studies produce many false positives." Put another way, before you start arguing about whether the result of a particular animal study is relevant to the effect of some chemical in humans you need to assess whether the study is even reliable in the first place. That's because the number of animal studies showing statistically significant results is far higher than would be expected by chance alone. In fact, it's almost double what you'd expect to find if hypotheses were being fairly tested in animals and the results invariably reported.

Interestingly, small sample size and financial interest seem to correlate with what appears to be p-hacking and/or selective publishing. Thus as we've said previously, you should assume the p-hacking scandal won't be limited to academia. The article is "Evaluation of Excess Significance Bias in Animal Studies of Neurological Diseases" and it sensibly ends:

In conclusion, the literature of animal studies on neurological disorders is probably subject to considerable bias. This does not mean that none of the observed associations in the literature are true.

 

 

 

Lawyers Have Learned To Distort Pharmacovigilance Signals

The goal of pharmacovigilance is to detect the subtle signal of a causal effect otherwise buried amongst the noise of all the maladies which by coincidence alone happened to manifest after a drug was taken. The FDA Adverse Event Reporting System (FAERS) gathers reports of these adverse events and presents them in a database that can be analyzed for just such signals. Since previously unknown causal associations are what are being sought there is no requirement that the reported adverse effect be demonstrably causal. To work properly the system merely requires accurate and unbiased reporting of as much adverse event data as possible.

You'd think health care professionals would be responsible for most of the reported adverse events and most of the time you'd be right. But zealous advocates (aka lawyers) can submit adverse reports too and when they do the resulting signal amplification is amazing (or appalling, depending on your perspective). Take the case of metoclopramide and tardive dyskinesia (TD) examined in:  "The Metoclopramide Black Box Warning for Tardive Dyskinesia: Effect on Clinical Practice, Adverse Event Reporting, and Prescription Drug Lawsuits".

For years the rate of TD per metoclopramide (Reglan) prescription bumped along at the bottom of the graph never rising above one half of one tenth of one percent. It is, or was, commonly prescribed in those suffering from gastroparesis. Then came a black box warning advising of the very rare but very serious side effect TD. Suddenly the TD rate took off like the Blue Angels at an airshow. What was behind it? "In the FAERS, only 3.6% of all the reports are from lawyers, whereas 70% of TD reports related to metoclopramide came from lawyers, resulting in a significant magnification of TD reports disproportionate to all other adverse reports." Interestingly, the authors using published legal opinions regarding TD/metoclopramide as a proxy for filings found that a wave of claims took off just before the wave of reports of TD. 

Distorting the metoclopramide/TD risk signal caused doctors and patients to overestimate the risk posed by metoclopramide. Often they chose instead domperidone which has not been approved by the FDA; unfortunately that "[puts] patients receiving this drug at potential risk for serious or even fatal arrhythmias."

Perhaps even more unsettling is the fact that as fewer and fewer prescriptions for metoclopramide were written the increase in the rate of TD got steeper and steeper. One hypothesis that readily emerges is that lawsuits drove the increase in TD cases rather than the other way around. If so, it points to another chronic problem in mass tort litigation - expert witnesses whose primary contribution is to dramatically over-diagnose ailments like TD.

The authors note that theirs is not the first study to find that lawyers are able to modulate the signals relied on for pharmacovigilance. It'll be interesting to see what if anything the FDA does about it.

Painting By Numbers

It's hard to argue with the decision of the U.S. Court of Appeals for the Seventh Circuit in Joann Schultz v. Akzo Nobel Paints, LLC; a benzene/AML (acute myelogenous leukemia) wrongful death claim filed by the wife of a painter. The opinion frames the question before the court as follows: Is the fact that plaintiff's oncology expert holds to the linear no-threshold model for carcinogens a sufficient reason under Daubert to exclude his opinion that plaintiff's estimated high benzene exposure was causative given published studies demonstrating an eleven-fold risk of AML among those similarly exposed? The obviously correct answer is "no" and so the court held. (Behold the power of framing!)

We haven't read any of the briefing and so don't know if the question as framed was really what the fight was all about (though we certainly hope there was more to it than just that) but we did read the entire opinion and noted a couple of things you might find of interest. One involves the fact that plaintiffs are finally getting why dose matters (and helps) another is the opportunity that arises when dose is calculated. The last is a lament about the common problem of conflating observational epi studies with the scientific method, casual causal inference via "differential diagnosis" and a pointer to a new tool you might find helpful.

Unlike most toxic tort plaintiffs Schultz had her expert perform a sophisticated dose reconstruction. From it the expert generated a cumulative dose estimate which was then compared to risk data drawn from epi studies plaintiff thought she could defend as scientifically sound. The comparison yielded a large increase in her deceased husband's risk of AML attributable to his benzene exposure. Now that's the way to do it. Better yet, it was essential to getting the summary judgment rendered against her reversed. How did she do it?

According to the appellate court Plaintiff reconstructed the decedent's benzene exposure "using Monte Carlo Analysis, a risk assessment model that accounts for variability and uncertainty in risk factors such as the likely variation in [decedent's] exposure to benzene during different periods and at different plants." The court then proceeded to write "[t]he U.S. Environmental Protection Agency (EPA) has endorsed this methodology as a reliable way to evaluate risk arising from environmental exposure." Sound good?

For the unwary it sounds as though Monte Carlo Analysis(/simulation/etc) is (a) some sort of mathematical equation that (b) generates reliable (and therefore presumed to be admissible) risk estimates while (c) accounting (somehow) for missing data, and it comes with (d) Uncle Sam's seal of approval. Unfortunately, it isn't so.

Rather than going into detail about Monte Carlo simulations and their promise and limitations let's briefly discuss what makes them invaluable when it comes to cross examining the other side's expert. It turns out that it's not some sort of mathematically deduced equation that "accounts for variability and uncertainty" in these Monte Carlo exercises - it's the expert who picks which variables matter along with the formula that expresses the pattern wherein the variables are expected to be found. In other words, the expert using a Monte Carlo method has translated her opinions into a mathematical language. So whereas an expert's deposition fought only in English typically yields dissembling about methods and sharp advocacy about results, a translation of the mathematical language of her model reveals her tactics and much of the other side's strategy. Words like "high" and "low" become e.g. 2 ppm and 0.01 ppm, "increasing" and "decreasing" become calculable slopes, and "it varied throughout his shift" becomes e.g. a power law - each suddenly vulnerable to an informed cross examination.

Yet while we cheered the use of (potentially) explicit and transparent models (we don't know if plaintiff's estimator divulged his spreadsheets and formulae) for estimating dose we groaned at an embarrassingly shabby standard for causal inference.

First there's the method: the differential diagnosis. Since plaintiff's expert ruled-in most of, or at least the important, "risk factors" for AML and then ruled-out everything besides benzene his opinion that benzene caused decedent's AML was deemed admissible. Ugh. Saying something is a risk factor is not the same thing as saying that it's a cause. That so-called risk factor epidemiology isn't science and isn't even likely to turn up real risks has been known for some time. Then there's the category error FAIL of ruling-in and -out of the set of causes things that aren't even causes. Next there's the "best of a bad lot" problem. If you don't have all the real causes "ruled-in" your diagnosis is iffy at best. If you don't have the most likely cause ruled-in then all you're likely doing is picking the least wrong cause. Since the cause of the vast majority of AML cases is unknown and as there was nothing to distinguish Schultz' AML from the thousands that arise spontaneously every year plaintiff's expert's failure to rule-out "whatever it is that causes 90% of all AMLs" should have been fatal to his differential diagnosis. (Note: it wouldn't however be fatal to plaintiff's case assuming the admissibility of her claim of an eleven-fold relative risk given decedent's estimated dose).

Second, there's the claim that because plaintiff's and defendant's epi studies share "the identical methodology - observing AML rates in populations exposed to benzene over time", "Rule 702 did not require, or even permit, the district court to choose between those two studies at the gatekeeping stage." The court would do well to read "The Scandal of Poor Epidemiological Research: Reporting guidelines are needed for observational epidemiology" before pronouncing such a rule. As one of the FDA EMDAC committee members said during the recent Avandia meeting, when it comes to evidence "observational studies are at the bottom of the totem pole." Courts should keep in mind the fact that small effects detected in such studies even though, and perhaps especially when, statistically significant (i.e. report a low p-value) are likely to be false - and that goes for the ones cited by defendants as well as plaintiffs.

If you're still harboring doubts about whether or not there really is an unfolding scandal involving observational epidemiology read the editor's choice, "Improving Observational Epidemiology" in the current edition of International Journal of Epidemiology. If you don't have free access to the entire paper this should encourage you to pay the price for it:

"The ability of observational datasets to generate spurious associations of non-genetic exposures and outcomes is extremely high, reflecting the correlated nature of many variables, and the temptation to publish such findings must rise as the P-values for the associations get smaller. The forces involved - the imperative to publish for a successful research career, the journal publishers' and editors' desire to publish material that gets cited to increase their profiles and the isolation of many epidemiologists working on small,  often non-contributory, studies - are strong. Perhaps epidemiology needs to re-define its training and knowledge base and build in subsequent accreditation routes to promote better standards of epidemiological professional practice. Very few epidemiology departments impose the discipline of laboratory daily log books in which every experiment and analysis is recorded to provide some verification of what was a priori and what was post hoc. Academics involved in 'handle-turning circular research', highlighted nearly a century ago by Paul de Kruif, and commented upon recently in this journal, really do need to find alternative pursuits."

Finally, if you're interested in uncovering abusive practices in observational epidemiology add p-hacking to your vocabulary (you know your meme has gone big time when it's on Urban Dictionary) and "P-Curve: A Key to the File Drawer" to your arsenal. A number of statisticians, alarmed at the realization that the tools of their trade have been used to gin up spurious science on an industrial scale, have developed new tools to detect it. Plaintiffs are turning such tools on drug studies and company-financed employee mortality studies. Meanwhile, in one case of which we're aware, defendants are using them with effect on a whole series of studies resting on nothing more than the curious coincidence that the reported p-values all fell between 0.044 and 0.05. Go figure. Literally.

Discretizations

A Duty To Mine Big Data

We had another jury trial and thus were offline for a few weeks but it didn't take long to dig up something I hope you'll find of interest. I ran across it in a recent opinion by the district court in the In Re Fosamax Product Liability Litigation. What is it? It's a duty to actively mine the FDA's data for a signal, a hint, that your approved pharmaceutical product might be associated with an adverse event.

As an initial matter I almost didn't get to the interesting part of In Re Fosamax Product Liability Litigation thanks to this eye-roller: "In applying the nine Bradford Hill factors, he [Dr. Cornell] reviewed Plaintiff's medical records from 1996 to present, the office notes and depositions of her treating physicians, and 'past and current medical literature on the topics of osteopenia, osteoporosis and their prevention and treatment with bisphosphonate drugs including alendronate'". Which was followed by: "The methodology Dr. Cornell used is sufficiently reliable because the Bradford Hill criteria are 'broadly accepted' in the scientific community 'for evaluating causation, and 'are so well established in epidemiological research". This business of giving expert witnesses a pass for doing nothing more than invoking the great epidemiologist's name and saying that their method consisted of peering at the evidence through the supposed lens of Hill's "criteria" is utterly appalling but it would take several paragraphs to explain and we've done it before so I'll leave it at that.

The portion worth pondering is found further down in the discussion of defendant's motion to exclude a different expert witness, Dr. Madigan. Madigan is a statistician from Columbia with an impressive resume. He was tasked with assessing "whether a signal of problematic oversuppression of bone turnover and associated [atypical femur fractures (AFF)] . . . existed for Fosamax, using industry standard pharmacovigilance techniques and data sources, and the adverse event terms selected by Merck to internally evaluate the same" and whether "the strength of that signal, if any, in comparison to the signal, if any, for such events in other products indicated for the prevention and treatment of osteoporosis". To identify and evaluate such a signal Madigan took several medical terms considered by the defendant to be possible indicators of "oversuppression of bone turnover" and/or AFF and, using a program called Qscan, ran them through FDA's Adverse Event Reporting System ("AERS") database looking for associations with Fosamax use.

The data revealed "the presence of a clear signal for oversuppression of bone turnover and associated atypical femur fracture events utilizing the terms selected by Merck for such analysis. By standard metrics of 'signal' detection, the signal is strong, consistent, and not ambiguous. Of perhaps greater concern, the signal was striking in comparison to that for other drugs indicated for the prevention and treatment of osteoporosis. As early as 2001-2002, the spontaneous report data for Fosamax provide signals for a number of indicators of suppression of bone turnover. For the comparator drugs, such signals either never appear or appear years later." As Qscan (or similar software) data mining is widely used (and often mandated in the pharmaceutical industry) and as peer-reviewed papers arising out of data generated by the software had been published the court concluded that "data mining in pharmacovigilance" is a reliable method. Fair enough.

The excitement comes in the next paragraph in which the court concludes that Madigan's testimony "fits", and so is relevant to, an issue in the case because it informs the question of whether the defendant should have warned of the later-perceived risk of AFF. The court thereby implicitly held, I think, that the defendant had a duty to mine the FDA's data as early as 2001-2002. In other words, the existence of powerful data mining tools capable of uncovering an early signal of a possible harm associated with defendant's product created a duty to use such a tool. Ultimately, that's a duty to discover any statistical association between your product and some harm in the FDA's (admittedly accessible) that might be causal and to thereafter warn about it; and it's a duty to mine not only your data but any data that might shed light on your product.

It's hard to know what to make of a duty to mine big data. Obviously it's a potential problem for defendants. "Should have known" is no longer what was reasonably knowable (by humans). Instead it's what was knowable given a duty to use "powerful data mining and signal detection capabilities" including "a powerful query-by-example module that allows users to mine and visualize data through inquiries utilizing multiple case data elements." We're already in the middle of something similar; a case in which the plaintiff is demanding to see all of the death certificates collected by our client's benefits department claiming that we had, or had assumed, a duty to look for mortality patterns among our workers that might suggest a work-related etiology - the idea being that the plaintiff's decedent had succumbed to one such workplace illness and we either knew or should have known from looking at the death certificates that danger had been lurking.

And of course there's the problem of experts coming in after the fact and running term after term after modified term in varying combinations until an association emerges. Working backwards they'll then construct a narrative about why the set of terms eventually founded to produce the association would have been the obvious choice at the time the data mining ought to have been done; and the failure to look for such an obvious potential association will thereafter be cast as willful ignorance.

Then there's the bane of defendants in all latent disease cases - the hindsight bias. It's the "knew it all along" fallacy that emerges when a jury is shown the picture on the box before they see the puzzle pieces inside. They can't thereafter imagine that the few anecdotes and case reports that constitute a handful of the pieces that make up the picture on the box could ever have suggested mere randomness. That means the ability to mine huge amounts of data will make signals easier to find while making it harder to mount a successful state-of-the-art defense any time Qscan can tease a signal from the data.

Data mining has led, and will lead to startling discoveries in the sciences. In the law it may well lead to startling liabilities - especially if defendants are made to pay for harms foreseeable only by the most powerful software available. Ponder that.

Those Who Switched From Saturated Animal Fats To Polyunsaturated Vegetable Oils May Have Made A Fatal Mistake

The publication of "Use of dietary linoleic acid for secondary prevention of coronary heart disease and death: evaluation of recovered data from the Sydney Diet Heart Study and updated meta-analysis" in the British Medical Journal should be sobering news for twenty-first century public health advocates. The results of the biggest and most thorough study thus far to investigate what happens when people switch from animal fats to vegetable oils couldn't be more clear. Those who heeded the advice to switch from saturated fats to polyunsaturated vegetable oils dramatically reduced their odds of living to see 2013.

What happened? Fifty years ago the medical community did an about-face (it had previously thought the evidence equivocal at best and so cautioned against jumping to conclusions) and instead went all in on polyunsaturated fats. It reasoned that since (a) cholesterol is associated with cardiovascular disease and (b) polyunsaturated fats reduce serum cholesterol levels, it inescapably followed that (c) changing people's diet from saturated fats to polyunsaturated fats would save a lot of lives. In 1984 Uncle Sam got involved - Time magazine reported on it in "Hold the Eggs and Butter" - and he made a big push for citizens to swap out animal fat in their diet for the vegetable variety and a great experiment on the American people was begun.

Assuming the inferences drawn from the Sydney Diet Heart Study (~60% increase in risk of death by switching from animal fats to vegetable oils) apply to Americans and considering the population at risk of death from cardiovascular disease (~37% given recent NHANES data) we can't think of any mass tort, or combination of mass torts, that has produced as much harm as the advice to change to a plant oil-based diet. Maybe some clever lawyer will find a way to use the tort system to quell the fervor of those armed only with a desire to do good and evidence of an association between something they don't like and something else they don't like. We won't be holding our breath. But we will be hoping that those seeking to impose lean body styles and low salt diets on their fellow countrymen take a good long look at the polyunsaturated vegetable oil data before they order their riders to sally forth on the next crusade.

Discretizations

Be careful with unsubstantiated claims!

An interesting and thoughtful comment on one of our recent posts from Dr. Marc-Andre Gagnon:

I believe your interpretation of the study is simply wrong. The authors refer to a study by Chan et al., published in BMJ, which showed that such differences between protocols and publications were prevalent in a sample of 70 clinical trials, 56 of which received industry funding. The quoted paper never said that non-industry funded studies showed as much difference (or bias) as industry funded studies, it simply made the case that we needed access to full data for every clinical trial to be sure that there are no bias in the reporting of results.

On the link on reporting bias in research and industry funding, one can find a Cochrane systematic review by Lungh et al. published on the topic two months ago.

Marc-André Gagnon

School of Public Policy and Administration, Carleton University

All the best,

Marc-André

Which precipitated an exchange wherein we asked about discrepancies and biases in non-industry funded data and got this equally thoughtful reply:

I think the topic is a very complex one since you need to take in consideration publication bias: journals like crispy results and dull ones are often unpublished. I think this type of bias applies in the same way to industry and non-industry funded studies. But this is not the same thing as reporting bias when it comes to the important question of industry funding and research outcome. A recent Cochrane systematic review showed that the problem is not an imaginary one: http://www.pharmaceuticalpolicy.ca/research/industry-sponsorship-and-research-outcome

The article analyzes the existing literature about reporting bias.

One thing is likely not in dispute - we're entering (or have entered) another age of empiricism in which data, and the inferences statistically drawn from it, presumptively trump even the cleverest idea if the data underpining that idea is biased, not honestly gathered or missing (in whole or in part). And that's a good thing. See also: "'In God we trust, all others (must) bring data’".

Thanks for the comments, Dr. Gagnon

 

Discretizations

Are Industry-Sponsored Clinical Trials Misleading Or Just As Sloppy As Those Done By Academics?

If you've read "Published clinical trials shown to be misleading", a story about a newly published article detailing discrepancies between a drug company's documents and the papers it published based on data drawn from those documents, and if you pondered this quote from one of the study's authors about their review of those internal company documents

[w]e could see all of the biases right in front of us all at once

then you probably think something really damning has been found. If you clicked through to the editorial published along with the article, "Getting More Generous with the Truth: Clinical Trial Reporting in 2013 and Beyond", you probably also agree that "[f]or many working in medical journal publishing, these results will sadly not be surprising" and would readily join the call to require greater transparency among drug companies conducting clinical trials of their products for safety and efficiency. Certainly that's been the common reaction seen across the various reports, blogposts and tweets about the article.

But if you actually read the article, "Differences in Reporting of Analyses in Internal Company Documents Versus Published Trial Reports: Comparisons in Industry-Sponsored Trials in Off-Label Uses of Gabapentin", you might reconsider. Take all of those biases right in front of the investigators. Presumably the nefarious drug company was at least clever enough to bias the results in the drug's favor, right? Consider this from the Results section:

...we did not assess whether disagreements in descriptions of types of analyses resulted in different participants being analyzed for efficacy and safety. We also did not examine the impact of the observed disagreements on the effectiveness of gabapentin for the indications specified in our study

Yet surely the discrepancies must be the unmistakable sign of the profit motive at work, distorting the truth, right? Well, there's also this

The discrepancies in description of types of analyses we observed in our study are not unique to trials sponsored by for-profit entities such as pharmaceutical companies. Using trial protocols obtained from institutional review boards and corresponding publications, previous research has shown that disagreements such as those we described in our study can be observed in trials with funding from both for-profit and not-for-profit entities.

So while many who read only the tweets and posts about the paper concluded that it's a familiar story of books being cooked to show a drug to be safe and effective when it's not, there was no such conclusion. It's sort of like a murder case with no body and no motive; and nobody missing. All that having been said, drug companies are held to a higher standard and when it comes to transparency they ought to assume that every mistake their employees make will eventually wind up on the front page of the NYTimes, or at least PLOS One.

 

Discretizations

The Annual Report to the Nation on the Status of Cancer (covering 1979 - 2009) is out. The good news is that overall death rates for some of the biggest killers continues to decline. The bad news is that deaths from cancers caused by infectious agents like HPV and hepatitis-C are up; doubling in some cases. Get your daughters, and sons, vaccinated.

It took several decades but the egg has now been cleared on the charges of murder by heart attack and stroke.

It also took decades to clear caffeinated coffee on all charges of murder by cancer. Now drinking 4+ cups per day has been shown to cut your risk of some cancers by half.

Many common food ingredients have been scientifically demonstrated to double or halve, and often double and halve, your risk of cancer. Click here to find out why.

The human breast milk microbiome is far more diverse than previously imagined and it's composition is impacted by factors such as obesity and Cesarean delivery. Could it be that her microbes, rather than a mother's genes, are to blame for her obese children? Could it be that it's the absence of certain breast microbes in mothers who underwent Cesarean delivery that's responsible for the increased risk of asthma in their children?

In the past, proponents of the Precautionary Principle have tended to deny that hormesis exists. As the evidence in favor of hormesis has gotten harder to ignore their narrative has changed to (1) it hasn't been shown to be the sort of general phenomenon that would permit an inference of hormesis for all toxins and (2) too little is known about the mechanism to permit its incorporation in to risk models.

Discretizations

Discretizations

Discretizations

Discretizations

Discretizations

Discretizations

Discretizations

Discretizations

The Best Causation Opinion of 2012

In fact, it may well be the best causation opinion of the last half century; and we're willing to bet it will be one of the most important causation opinions going forward. Dixon v. Ford Motor Company nails substantial factor causation and in the process proves that there are still plenty of judges willing to think deeply about causal inference, about uncertainty and about the limits of liability in a world of inevitable risk. Ultimately, the court held that risk is the measure of legal causation in those cases where causal inferences are made probabilistically, and therefore when a causation expert opines that a putative cause was a "substantial factor" without saying how much risk it imparted she fails to answer the question she was called to give; and so has nothing helpful to say about the matter.

Dixon is another tragic mesothelioma case in which a defendant, having contributed to the victim's cumulative asbestos exposure something between nothing and next to nothing, was hammered by the jury. It then appealed, after taking a few judicial deductions, a $3+ million judgment. As usual the defendant complained that plaintiff's causation expert (this time an epidemiologist) ought not to have been allowed to testify because she had "extrapolated downward" from the known segment of the asbestos/mesothelioma dose/response curve to an area in which hard data was lacking and thus could not reliably say that the defendant's small contribution to the victim's dose was a substantial causative factor. 1) That's not plaintiffs' game these days, as we said in Small Glasses; but more importantly, 2) as the court made plain in Dixon, "substantial factor" in such cases isn't about causation, it's about risk.

"'Substantiality' is a legal concept and not an objective property testable by the scientific method", wrote the Dixon court. The question thus required to get at the essence of substantial factor causation in a dose/response disease case is not "was it a big cause or a little cause?" It's not even "was it really and truly a cause?" The first question is nonsensical and the second is unanswerable. Instead, the question is: "assuming it was a cause, was the risk it imparted prospectively of such a degree as to justly warrant the imposition of liability?" Risk, said the court, is thereby the measure of (legal) causation and proving that the risk imparted was substantial is plaintiff's burden. Showing that defendant imparted a substantial degree of risk, according to the Dixon court, is thereby what is required to bridge the "analytical gap" between "asbestos causes mesothelioma" and "plaintiff's exposure to defendant's asbestos caused her mesothelioma".

Accordingly, the court held that when all plaintiff's causation expert could say about causation was that "every exposure to asbestos is a substantial contributing cause" the only thing she wound up saying about the risk given plaintiff by Ford was that it was "more than nothing." "For obvious reasons an infinitesimal change in risk cannot suffice to maintain a cause of action in tort". Thus, the opinion of plaintiff's expert "merely implied that there was some non-zero probability that [plaintiff] was exposed to asbestos from Ford's product, and that this resulted in some non-zero increase in her risk of contracting mesothelioma. As such, [the conclusion of plaintiff's expert] that the risk and probability of causation was 'substantial' provided the jury with nothing more than her subjective opinion of 'responsibility,' not scientific evidence of causation."

So how does a plaintiff establish a substantial risk? Not by proving the exact dose and thus exact risk incurred.The Dixon court disposed of the straw man argument that requiring plaintiff to show risk (and thereby dose) quantitatively demands an impossible degree of certainty and precision. Such a rule, wrote the court, would obviously be "folly". Rather, all plaintiff is being asked to do is to "estimate exposure and risk with reasonable scientific or medical certainty." Once such an estimate is made the jury can decide whether it is sound and if so whether the risk imparted was substantial - subject to the law's requirement that it be more than de minimis.

That's exactly what we've been arguing since a client was kind enough to let us write an amicus in Borg-Warner v. Flores. Make the plaintiff say what dose, and thereby what risk, was given. Why? The effect of plaintiff's estimate of dose in one of the first post-Borg-Warner cases should suffice for an answer. Using their expert's dose estimate the calculated risk of death posed by our client's product was demonstrated to have been less than 1 in six billion; that's equivalent to the risk of death (from cancer caused by radon) imparted by spending just fifteen minutes in a building constructed of brick or stone. Such calculations and comparative risk exercises serve to vividly demonstrate both the de minimis nature of the risks imparted and the absurdity of imposing liability for them.

Discretizations

PLoS Medicine is Publishing An Attack On "Big Food"

A new series in PLoS Medicine says we're going through another epidemiologic transition; this time it's a "nutrition transition", from a simple traditional diet to a highly processed food diet "resulting in a stark and sick irony: one billion people on the planet are hungry while two billion are obese or overweight". Can you guess who gets the blame? Can you guess what playbook they use? Here are some hints:

"In contrast to the actions of Big Tobacco, soda industry CSR initiatives are explicitly and aggressively profit-seeking."  CSR = corporate social responsibility

Neoliberal policies, including the opening of markets to trade and foreign investment, create environments that are conducive to the widespread distribution of unhealthy commodities by multinational firms.

Big Food attains profit by expanding markets to reach more people, increasing people's sense of hunger so that they buy more food, and increasing profit margins through encouraging consumption of products with higher price/cost surpluses

So are we in the midst of yet another epidemiologic transition? The last one was a bust. It turned out that we never really left the age of infectious diseases. Our bet is that the war on "Big Food" may generate fees but will do little to alleviate either hunger or the obesity epidemic.

IARC's Paper on Diesel Exhaust, Gasoline Exhaust and Nitroarenes Is Out

See "Carcinogenicity of diesel-engine and gasoline-engine exhausts and some nitroarenes" at Lancet Oncology.

Diesel exhaust is elevated to Group 1 while gasoline exhaust is 2B. Most of the nitroarenes are 2B though 6-nitrochrysene was elevated to 2A. Clearly the thinking is that exhaust particulates from which tars can be extracted are what's driving the increased lung cancer risk as the exhaust gases alone showed little to no sign of genotoxicity. Regarding the epidemiology, some found cumulative dose to be associated with risk while others found length of exposure, irrespective of dose, to be driving the association. Only one found a big increase in risk and that was only among those most exposed.

Expect to see more lung cancer claims by truckers, railroad engineers and others routinely exposed to diesel exhaust in their work.

IARC Moves Diesel Engine Exhaust To Group 1

Referencing two articles published in The Journal of the National Cancer Institute earlier this year, which examined the same cohort of 12,315 miners and found small though statistically significant increases in lung cancer risk that more or less varied with retrospectively estimated exposures, and the unanimous conclusion of its panel after a week of meetings that there was sufficient evidence to conclude that diesel exhaust probably causes lung cancer, IARC today elevated diesel exhaust to Group 1 status along with asbestos, tobacco, radon, etc. The argument for the move will be published online Friday at Lancet Oncology so we'll hold our thoughts until we've had a chance to read it.

You can read IARC's press release here and find the two articles here and here.

While you wait for Friday's Lancet Oncology ponder the implications of this sentence from the press release - it manages quite nicely to illuminate IARC's philosophy regarding regulations:

There is a strong interplay between standards and technology - standards drive technology and new technology enables more stringent standards.

Discretizations

Small Glasses

The Betz v. Pneumo Abex commentary has generally sounded the following themes:

(a) The "novelty" threshold requirement for a Frye hearing has a broader meaning than previously thought. Henceforth, "a Frye hearing is warranted when a trial judge has articulable grounds to believe that an expert witness has not applied accepted scientific methodology in a conventional fashion in reaching his or her conclusion."

(b) The trial judge, who "was unable to discern a coherent methodology supporting the notion that every single fiber from among, potentially, millions is substantially causative of disease", did not abuse his discretion by excluding the "every fiber" causal theory of Dr. Maddox.

(c) Maddox' opinion that "every fiber was a substantial factor in causing plaintiff's mesothelioma" is riven by an "irreconcilable conflict with itself" because "one cannot simultaneously maintain that a single fiber among millions is substantially causative, while also conceding that a disease is dose responsive."

(d) Maddox reached his conclusion about causation not by his claimed "series of 'small bridges'" but rather by improperly extrapolating from known portions of the asbestos/mesothelioma dose-response curve to find causation at much lower levels - levels never demonstrated by epidemiological studies to be associated with the disease.

We're glad, of course, that the court found that novelty doesn't wear off expert opinions (i.e. they don't become immune to scrutiny under Frye) just because they've been peddled to many juries for many years. And we're similarly pleased so see shot down the untestable/unverifiable/unfalsifiable claim that a single fiber not only might have caused, but was in fact a substantial cause, of a given case of mesothelioma. But we're afraid the court missed the subtle game Maddox was really playing.

He wasn't extrapolating down. He wasn't playing the one-hit, linear no-threshold, conflate causation with risk game (though he does in other jurisdictions). And his causal model can account for both a single fiber being a necessary cause of a given plaintiff's mesothelioma and a correlation between dose and disease without any tension. Best of all, it's perfect for exploiting jurisdictions with naive liability attribution schemes such as Pennsylvania's at the time Maddox's opinion was rendered (Pennsylvania finally overhauled its joint and several liability scheme in 2011).

To understand what's up think about the following opinions of Maddox:

1) "it is the total and cumulative exposure that should be considered for causation purposes."

2) "Cumulative exposure, on a probability basis, should thus be considered the main criterion for the attribution of a substantial contribution by asbestos to lung cancer risk."

and,

3) "[T]he more common analogy that has been used is the example of a glass of water. One drops marbles into the glass of water until the water finally overflows from the glass. Is it the first marble or the last marble that causes the glass to overflow? Well, both, all of them. The marbles cause the glass to overflow. That's a cumulative effect."

So it's clearly not a one-hit model that he's proposing; and for good reason. As more and more courts have become savvy about risk it's been harder and harder for plaintiffs to survive the objection that a de minimis exposure presents a de minimis risk and thus cannot by necessity be a "substantial factor". That's what the Pennsylvania Supreme Court was getting at in Betz when it wondered how "if all Dr. Maddox could say is that a risk attaches to a single asbestos fiber - that he could also say that such risk is substantial when the test plaintiffs may have been (and likely were) exposed to millions of other fibers from other sources including background exposure."

The point is made explicitly a few pages later when the court refers back to its discussion in Gregg about the plaintiff's proof problem in such cases:

"... we do not believe that it is a viable solution to indulge in a fiction that each and every exposure to asbestos, no matter how minimal in relation to other exposures, implicates a fact issue concerning substantial-factor causation in every "direct-evidence" case. The result, in our view, is to subject defendants to full joint-and-several liability for injuries and fatalities in the absence of any reasonably developed scientific reasoning that would support the conclusion that the product sold by the defendant was a substantial factor in causing the harm."

So what sort of model makes each asbestos fiber a substantial factor in causing mesothelioma and yet is consistent with a dose-response relationship between asbestos and mesothelioma? It's something we're seeing more and more often. It's the idea that each person has, to apply Maddox' metaphor from above, her or his own unique water glass of a defense mechanism. Some people's defense capacities are bigger and some are smaller and some are microscopic. But whatever the size, once they're overwhelmed mesothelioma or leukemia or whatever ensue. Thus, when Maddox says that every fiber to which plaintiff was exposed contributed to his risk and "once an individual develops a mesothelioma, the risk becomes the cause" what he's really saying is that though some marbles dropped in the glass might be bigger than others (e.g. crocidolite > amosite > chrysotile) and some might have been added with thimbles while others with coal shovels the only thing that really matters is whether or not the body's defenses were breached. Once they are the glass overflows, the mesothelioma develops and every last fiber was literally a necessary "but for" cause of the disease.

So what does the dose-response curve for asbestos and mesothelioma actually represent in such a model? It describes the outcomes of varying abilities to withstand asbestos across varying levels of exposure to asbestos. A lower defense capacity (small glass) requires less asbestos to trigger mesothelioma; higher defenses require more, etc. Think of the data points on the curve then as small glasses that have overflowed. What such a model suggests is a human population that mostly has large glasses {which fits the observation that even the most heavily exposed workers have less than a 1 in 10 chance of developing mesothelioma); a small portion has small glasses; and a tiny portion have smaller glasses still - all the way down until you get to those rare but unlucky souls who have no defense whatsoever. For them the background exposure dose doesn't increase their risk since it's already 1.

It's an especially clever model for for litigation because:

a) every fiber is a necessary element of a sufficient causal set (NESS) so causation is a snap;

b) it escapes the no duty problem posed by de minimis exposures in states that view substantial factor causation from a risk/foreseeability perspective;

c) it turns modern victims into eggshell plaintiffs and so makes every case that much more foreseeable; and,

d) because it's a Landers v. East Texas Salt Water Disposal Co sort of cumulative causal claim it's especially dangerous in states with joint and several liability schemes.

The problem is that there's no evidence for the model. It rests solely on the following two generalizations (that do little more than highlight our ignorance of the mechanisms whereby asbestos causes mesothelioma) and the conclusion drawn from them:

Mesothelioma is caused by exposure to asbestos fibers

A victim's asbestos exposure consists of every fiber inhaled

Therefore, a victim's mesothelioma was caused by every asbestos fiber inhaled

As for the equally evidence-free small glasses theory it goes like this:

Not everyone exposed to asbestos develops mesothelioma

But every victim's mesothelioma was caused by her or his asbestos exposure

And every victim has had a different asbestos exposure

Therefore something about the victim (small glasses) determines the outcome of exposure

If you haven't seen it yet it's surely on its way to a courtroom near you. Our advice? Attack the premises - they're made of clouds.

 

 

 

 

Discretizations

 "'Hunches', even if held by experts, are not scientific knowledge."

Banning plastic grocery bags might be a very bad idea.

There's no such thing as negligent failure to generate the same hypothesis as plaintiff's future experts.

There's a very slight increased risk of lung cancer among Carolina chrysotile textile workers with 100 f-yr/ml exposure.

Apparently, the Restatement (Third) of Torts, Liability for Physical and Emotional harm, section 7's exception permitting no-duty rules when "relatively clear, categorical, bright-line rules of law" can be promulgated is big enough to drive the very big take-home truck through it. We predict dyspepsia for Michael Green. See Mary Campbell v. The Ford Motor Company.

Discretizations

Discretizations

Bending the Dose Response Curve

The linear no-threshold model of dose-response meant that plaintiffs could continue to prevail on toxic tort claims even though their exposures had occurred in the modern era and thus were tiny fractions of those that led to epidemics in years past. Either courts permitted plaintiffs to rely on a one molecule / one particle theory of causation (consistent with the view that some risk is associated with a single molecule or particle) or they allowed plaintiffs to conflate causation with risk.

Eventually some courts began to grasp the absurdity that follows from basing proximate cause on a "one-hit" model in a world of trillions of hits while others began to take notice of the fact that despite probing larger and larger populations with low exposures epidemiology was unable to verify the linear no-threshold model for numerous diseases; thereby suggesting that there is indeed a threshold for diseases including leukemia (a new case making the latter point is Schultz v. Glidden Company.) Meanwhile we have argued that the old cases got it right - that causation in an individual toxic tort case is unfathomable and that the most sensible approach is to estimate the risk imparted (e.g. by a single molecule); to ask why it makes sense to impose liability for creating a 1:1,000,000,000,000,000,000 chance of harm; and, further asking why it wouldn't make sense to impose liability for a 1:100,000 or greater risk.

But all of that assumes risk goes to zero or at least continues to decrease as exposure is reduced below previously measured levels. If that assumption is false, if risk starts heading back up as exposure goes down, especially if unpredictably so, then all bets are off. We will have entered another period of great uncertainty, And it's in such times that toxic tort claims flourish. The horsemen of this new age of uncertainty have published a review paper on the topic and if you want to understand what's coming, why it's pitch perfect for the health and wellness movement and why what happened to BPA will be repeated again and again for other chemicals until some new way is established to either verify or refute their claim that dose doesn't make the poison you need to read it:  "Hormones and Endocrine-Disrupting Chemicals: Low-Dose Effects and Nonmonotonic Dose Responses"

Discretizations

When it comes to lung cancer, keep your chickens close but keep your flock closer.

"They've internalized their own confusion": watch this video if you want to understand the difference between humble falsificationist evidence-based science and the authoritarian "because I say so" approach adopted by the 1st Circuit in Milward.

Antibiotic use in children less than six months old is strongly associated with the subsequent development of asthma, especially in those without a family history of asthma, and the greater the intake of antibiotics the greater the odds of developing asthma.

Attacks on randomized controlled trials (RCTs) miss the point.

A new paradigm for cancer prevention, indeed.

 

Clostridium Difficile Infections Are At Historic Highs

The CDC has a series of reports out about the growing problem of C. difficile infections contracted via health care-associated activities. The number of deaths attributable to C. difficile has increased more than seven-fold in a little over a decade and now stands at 14,000 annually.

The reports twist together a few threads we've been following for awhile now. First, chronic and often unnecessary antibiotic use in patients disrupts their normal gut microflora and provides an opportunity for C. difficile to gain a foothold and thereafter thrive. Second, these infections aren't just hospital-acquired, or nosocomial, anymore. Infected patients, especially those with diarrhea and on antibiotics transmit the pathogen to surfaces in clinics, nursing homes and doctors' offices and staff thereafter transmit it to other patients; thus they're healthcare-associated infections (HAIs). Finally, the infection is preventable.

The CDC makes the following recommendations to clinicians:

1. Prescribe and use antibiotics carefully. About 50% of all antibiotics given are not needed, unnecessarily raising the risk of C. difficile infections.

2. Test for C. difficile when patients have diarrhea while on antibiotics or within several months of taking them.

3. Isolate patients with C. difficile immediately.

4. Wear gloves and gowns when treating patients with C. difficile, even during short visits. Hand sanitizer does not kill C. difficile, and hand washing may not be sufficient.

5. Clean room surfaces with bleach or another EPA-approved, spore-killing disinfectant after a patient with C. difficile has been treated there.

6. When a patient transfers, notify the new facility if the patient has a C. difficile infection.

If you have a C. difficile case you might want to see how the healthcare provider filled out this form.

For more information try the following links:

CDC's March 2012 "Vitalsigns"

Centers for Medicare & Medicaid Services (CMS) Guidelines for Reducing the Spread of C. difficile.

Finally, there's a brand new research paper from the CDC which discusses the finding that C. difficile is an emergent disease affecting patients recently exposed to antimicrobial drugs.

 

April's "Gut" Has Some Very Interesting Articles.

Have a look at the cover and you'll see why double-dipping chip dippers are frowned upon. Elsewhere you'll find evidence that that if you want to stay lean or if you want to avoid type II diabetes you need to hold down the ratio of Firmicutes to Bacteroidetes that live in your gut. If you're looking for evidence that keeping H. pylori at bay keeps gastric MALT lymphoma away you'll find that too. There's also evidence that pancreatic cancer is associated with just a couple of bacteria found inthe saliva of those at risk of the deadly disease. Here's the link: Gut

Discretizations

There's Something About Nevada

After getting hammered in quick succession for $500 million, $162 million and $90 million worth of punitive damages in three cases involving a total of just five plaintiffs, Teva settled its Nevada hepatitis C (HCV) / Propofol claims.  The price tag was just past a quarter billion dollars. 

So what happened?  Teva's generic Propofol wasn't defective.  It didn't leave the factory contaminated with HCV.  Instead the claims arose out of the following findings of the CDC:

Inappropriate reuse of syringes on individual persons and use of medication vials intended for single-person use on multiple persons was identified through direct observation of infection-control practices at clinic A (Desert Shadow Endoscopy Center). Specifically, a clean needle and syringe were used to draw medication from a single-use vial of propofol, a short-acting intravenous anesthetic agent. The medication was injected directly through an intravenous catheter into the patient's arm. If a patient required more sedation, the needle was removed from the syringe and replaced with a new needle; the new needle with the old syringe was used to draw more medication. Backflow from the patient's intravenous catheter or from needle removal might have contaminated the syringe with HCV and subsequently contaminated the vial. Medication remaining in the vial was used to sedate the next patient.

and the final report of the Southern Nevada Health District which includes the following graphic demonstrating how the clinic contaminated the Propofol:

The vials were clearly labeled for single use and the warning was further emphasized when reports surfaced that some were being used for multiple doses.  Discovery apparently revealed that the larger vial size helped reduce manufacturing costs. 

Okay, so Teva was found liable for not guarding against the (allegedly) foreseeable blatant disregard of its warning labels and of standard vaccination administration safety procedures by the endoscopy clinic.  Nevada seemingly has an expansive view of duty and its jurors a limitless view of foreseeability.  But what explains the repeated mega punitive verdicts?  Every mass tort has its outliers but we've not seen anything like this.  If we hear any scuttlebutt we'll let you know.

Along the way it certainly looked like the cases were going well for Teva.  For example, it made some good law at the Nevada Supreme Court which we've mentioned previously.  Specifically the court recognized that a rule preventing a defendant from discussing other possible causes (unless its experts are willing to testify that such alternate cause(s) were, in reasonable medical probability, the cause in fact of plaintiff's illness) would work a great injustice.  As we've said before the effect of such a rule would be to admit the most speculative claims of causation while excluding the most reasoned and sober assessments of causal attribution.

Why did Teva settle?  Again, we're not privy to any inside information but hazard the following guess.  The recent tendency of the Nevada Supreme Court to affirm eye-popping jackpot verdicts weighed heavily in Teva's analysis.  Usually when you get a huge judgment before a supreme court, as Teva had here, you're pretty confident that at the very least the outlier award will be substantially reduced.  Yet in Nevada these days such payouts are quite likely to be affirmed. In any event we're betting that the Nevada civil courts will be busy and that tort claims arising out of the spread of pathogenic bacteria, viruses and prions will increasingly be the focus of the plaintiffs' bar.

Discretizations

Discretizations

Discretizations

The conjunction paradox isn't one and doesn't require new standards of proof; rather, it's evidence that the asymmetry in the law markedly favors false positives, whatever some may say.

Cancer-causing H. pylori might find its way into you by hitching a ride in food borne yeasts.

Blame for obesity: BPA vs BUG.

The evidence that hormone replacement therapy "causes" breast cancer isn't as strong as you think.

Occupational exposure to animals: evidence for a causal link with cancer mounts.

Discretizations

Discretizations

Our Lives Are Better Left To Chance

The science of epidemiology is coming to grips with the following facts: a) no "big" risk factors have been discovered in quite awhile and the future looks to consist mainly in attempting to uncover tiny ones; b) even when big risks in a population are identified the knowledge does not translate into making sound predictions about individuals; because c) chance, or luck, plays a huge role in whether or not we'll get sick, whatever our exposures. That's my take on an article you need to read if you're at all interested in epidemiology - it's "Epidemiology, Epigenetics and the 'Gloomy Prospect': Embracing Randomness in Population Health Research and Practice".

Below are a few lines from the exceptionally well written piece to whet your appetite:

"We should embrace the effects of chance, rather than pretend to be able to discipline them."

"The pioneering epigeneticist Robin Holliday points out that it is commonly stated that disease is either genetic or environmental, when in reality stochastic events are equally important."

"Epidemiological inference is to the group, not to the individual"

"... the determinants of the incidence rate experienced by a population may explain little of the variation in risk between individuals within the population."

"Chance (at one level) and near necessity (at another) may be the only certainty in attempting to understand epidemiological - and many other - processes."

"At a group level, the underlying social causes of IHD (ischemic heart disease) could be social and political structure, sequentially mediated through free trade in toxic microenvironments, in health-related behaviours, and in the elevated body mass index, blood pressure, serum cholesterol, glucose and insulin. At an individual level, it is mostly genes and chance."

"... in terms of public health policy, we should target the modifiable causes of disease that heritability and shared environment tell us about. This must be at the group level, however, and we should do so without pretending to understand individual-level risk, or misrepresent population level data (smokers die earlier on average) as individual level events (each smoker shortens her or his life."

Pay special attention to the section on "lay epidemiology". The narratives we create about our lives, cast in terms of fate, or chance, turn out to be a lot closer to the mark than most epidemiologists have been willing to admit.

 h/t Garth Brooks

"... It is Unacceptably Easy to Publish "Statistically Significant" Evidence Consistent with Any Hypothesis"

Want to look and feel younger?  Well, there's a properly done study, statistically significant at p < .05, showing that people who listen to The Beatles "When I'm Sixty-Four" actually became a year and a half younger!  Far fetched?  Sure, but no more so than umpteen conclusions published in the scientific literature every day purporting to establish some causal connection based on nothing more than a statistical analysis of a series of observations.  That's the point demonstrated conclusively in False-Positive Psychology: Undisclosed Flexibility in Data Collection and Analysis Allows Presenting Anything as Significant.

Though a paper may validly claim that the likelihood of the reported causal connection being due to chance alone is 5% or less (i.e. p < .05), "False-Positive Psychology" demonstrates that researchers free to modify as few as four variables (such as the number of observations to be made, sorting those observed by gender or stratifying outcomes) more likely than not have "discovered" a causal association that doesn't exist.  The harm done by publishing such false-positives are obvious.  As the authors put it:

 "First, once they appear in the literature, false positives are particularly persistent. Because null results have many possible causes, failures to replicate previous findings are never conclusive. Furthermore, because it is uncommon for prestigious journals to publish null findings or exact replications, researchers have little incentive to even attempt them. Second, false positives waste resources: They inspire investment in fruitless research programs and can lead to ineffective policy changes. Finally, a field known for publishing false positives risks losing its credibility."

To vaccinate against infecting the scientific literature with false-positives the authors conclude with suggestions similar to those we've seen elsewhere in efforts to promote evidence-based science.  At the heart of the suggested approach is transparency from the moment the experiment is conceived all the way through publication.  There are six for authors and four for reviewers; be sure to read them all.

Discretizations

The NYTimes says extra sodium threatens all but a new study says normal people just excrete it

The XMRV / CFS (Chronic Fatigue Syndrome) fiasco is simply an example of how science works

European study finds increased risk of death (CVD and heart failure) for Avandia vs pioglitazone beginning shortly after initiation, and disappearing upon discontinuation, of treatment

Can Vitamin D prevent fractures and even cancer? The evidence is mostly all over the board

Paternal, but not maternal, smoking is associated with an increased risk of childhood leukemia

Discretizations

Discretizations

Discretizations

Discretizations

The Texas Sharpshooter Goes Free Range

Back in 1987 right after I'd gotten out of law school and passed the bar one of the first things I worked on was objecting to a request by plaintiff's counsel for all of the death certificates gathered by the University of Texas School of Public Health for its study of our client's refinery. Plaintiffs' retained epidemiologist wanted to do his own "unbiased" study, he claimed.

"Have you heard of the Texas Sharpshooter?", asked the school's biostatistician when I met with him for the first time. No, I hadn't. But I would soon understand that when you get to draw your target after you see where your shots have struck, your accuracy improves dramatically. And that was exactly what happened when plaintiffs' expert did his study. Suddenly big risks appeared where none had been found by other investigators.

The problem, of course, was that the expert's hypothesis was generated by the litigation, and not by scientific curiosity. He knew who had developed leukemia and so, after reviewing payroll records and depositions, honed in on the inevitable commonalities among the afflicted workers. Thus he discovered what tasks must, in his mind, have lead to the causative exposures; and, sure enough, his unpublished study showed that that there was a big increase in leukemia among those with work histories like our plaintiffs.

Yet even in Beaumont, even in the bad old days, the judge who usually gave us fits (admitting the Sumner Simpson papers against defendants who'd never heard of them, much less participated in the alleged conspiracy) recognized the fallacy embedded in the claims of plaintiffs' expert and excluded his ex post epi study (though he had let plaintiffs get the death certificates). You can always find something that events have in common after they've occurred, the court held. The trick, and the real test of a scientific theory, is whether it can predict those events before the test is run (or the Texas sharpshooter's shots are fired).

According to the scientific method a sound test is one designed to test a theory's predictions and not just to verify the investigator's preconceived notions. Sadly, that distinction was overlooked by the New York appellate court in Nonnon II. Nevertheless, all surely cannot be lost when such things go better in Illinois. The very point missed by the court in Nonnon II was seized upon by an Illinois trial court. It examined the findings of the same expert doing the very same sort of ex post epi study who arrived at the same conclusion (which again flew in the face of objective, blinded studies done by the state) and, deciding enough was enough, threw out plaintiffs' claims. Looking at the two rulings it appears we haven't come very far in the last 25 years. On one side is the emotive power of claims fanned by fears of childhood leukemia clusters and on the other the cold hard facts and a whole lot of uncertainty. Not much of anything has changed I suppose.

Note to readers: We're coming up on 800 posts and I've just now learned that I routinely fail to check the "Send Notification" box upon posting. Thus those of you who've signed up haven't gotten notifications of the vast majority of our musings. I apologize and promise to check the box henceforth. h/t Laura K. Deen.

 

Discretizations

Discretizations

Discretizations

Discretizations

Havner: Now We're Really Confused

The Texas Supreme Court just decided Merck v. Garza. The relatively short opinion rolls along (1) reaffirming Havner; (2) apparently adding the further requirement of a second well done epidemiological study "statistically significant at the 95% confidence level" that shows a doubling of risk; (3) rejecting the "totality of the evidence" ipse dixit of plaintiff's expert; but then suddenly (4) utterly confounding us by holding "when parties attempt to prove general causation using epidemiological evidence, a threshold requirement of reliability is that the evidence demonstrate a statistically significant doubling of the risk". What?!

The whole purpose of the "doubling of the risk" requirement had been, we thought, to ensure that when a plaintiff has nothing but probabilistic evidence such evidence must actually support a "more likely than not" causal inference as to her specific illness. There are numerous agents that produce small effects (i.e. relative risks less than 2.0) and which are nevertheless unquestionably causative of human disease. Hopefully, the court meant "specific" where it wrote "general" regarding risk doubling.

Yet there's another problem on the very next page. Apparently courts are now to "examine the design and execution of epidemiological studies using factors like the Bradford Hill criteria to reveal any biases that might have skewed the results of the study." Again: What?! We thought (and we're pretty sure we're right) that Hill's list of factors were his way of assessing a given claim of general causation. And anyway, that's not how you look for bias. This is how you look for bias: "Excess Significance Bias in the Literature on Brain Volume Abnormalities".

In sum we liked, of course, the court's conclusion that when each piece of plaintiff's supposedly supportive evidence is flawed "a plaintiff cannot prove causation by presenting different types of unreliable evidence." Yet, recognizing that causal inference is hard (nearly maddening sometimes) and that statistical inference is complicated and counterintuitive, we wish the court had done a better job on this one. The deviations from standard analysis will only support those who complain that the current court is "merely results oriented".

Speaking of Dusty Death

 ... there are lots of new papers associating various sorts of dusts with cancer, pneumonia and cardiovascular disease. We've previously discussed the correlation between endotoxins and a reduced risk of lung cancer but how to square those studies with ones like "Occupational Exposure to Organic Dust Increases Lung Cancer Risk in the General Population" (which identifies exposure to dust from "microbial, plant or animal" sources)?

On the inorganic dust front there's "Increased Mortality From Infectious Pneumonia After Occupational Exposure to Inorganic Dust, Metal Fumes and Chemicals". It's a study of 320,143 workers in the construction industry that finds a fairly large increase in risk of death from pneumonia among those exposed to a mixture of inorganic dusts yet the opposite outcome for those exposed just to one sort of dust.

Finally, on the pm2.5 front there's "The Effect of Particle Size on Cardiovascular Disorders - The Smaller the Worse", focusing on, obviously, size rather than substance; "The Effects of Particulate Matter Sources on Daily Mortality: A Case-Crossover Study of Barcelona, Spain" finding again that the observed correlation is a matter of size over substance when it comes to an increased risk of cardiovascular disease; and, evidence that the observed association is not, perhaps, the result of confounding whereby people more prone to cardiovascular disease (for which socioeconomic status is a big risk factor) are more likely to live near sources of pm2.5 can be found in "Particulate Air Pollution and Socioeconomic Position in Rural and Urban Areas of the Northeastern United States".

 

Erionite in North Dakota

Mesothelioma plaintiff lawyers will no longer be able to mockingly ask their clients "Have you ever been to Cappadocia, Turkey?" That's because it turns out there's plenty of erionite in the USA. For more info see, just  published in the Proceedings of the National Academy of Sciences of the United States of America: "Erionite Exposure in North Dakota and Turkish Villages With Mesothelioma".

This Week Cell Phones Don't Cause Brain Cancer

The ICNIRP (International Commission on Non-Ionizing Radiation Protection) Standing Committee on Epidemiology recently analyzed published research related to cell phones and brain cancer. This review concludes that evidence from a growing number of studies does not support the theory that cell phones raise the risk of brain cancer.

It's interesting to note that early on D. Savitz, a member of the ICNIRP Standing Committee on Epidemiology, was in the EMF causes childhood leukemia camp. In 1993 with the accumulation of additional data Savitz was able to concluded that “the evidence falls short of demonstrating a causal association between electric and magnetic fields and cancer.”

In 2004 the same members of the ICNIRP Standing Committee on Epidemiology published a detailed review of epidemiological studies of health effects from exposure to radio waves. The reviewers concluded that "results of these studies to date give no consistent or convincing evidence of a causal relation between RF exposure and any adverse health effect."

In 2009 the ICNIRP Standing Committee on Epidemiology published an update of their 2004 review. They noted that the number of papers on this topic had grown since 2004, but concluded that the available data does not suggest a causal association between mobile phone use and fast growing tumours in the brain such as malignant glioma. The similar absence of an association for slow growing tumours such as meningioma and acoustic neuroma is far less conclusive because the period of observation is simply too short.

With ICNIRP’s caveat “the possibility of a small or a longer term effect cannot be ruled out", I’m sure this isn't the end of the story. We’ll keep you posted.

Discretizations

A risk-based tax on rubber duckies?

Tumors are superorganisms.

Kentucky adopts the economic loss rule. (An especially well written opinion btw).

To be injured, you have to be injured. (Wis Ct of Appeals: another well-reasoned opinion worth reading)

Roggli shows that brake repair workers got their meso from commercial amphiboles.

Discretizations

Discretizations

Discretizations

Low-Salt Mandates: Who Pays for the Damages?

By now you've probably read "Low-Salt Diet Ineffective, Study Finds. Disagreement Abounds" by Gina Kolata of the NYTimes. She's reporting on "Fatal and Nonfatal Outcomes, Incidence of Hypertension, and Blood Pressure Changes in Relation to Urinary Sodium Excretion" published this week in JAMA. The big news is that lowering salt in healthy adults was associated with an increased risk of death from cardiovascular disease (CVD). Meanwhile, those subjects in the same study on significantly higher sodium diets were much less likely than those on low-sodium diets to die over the years of follow-up (a median of 7.9yrs for the 3681 members of the cohort).

What you might not have read though are the numerous articles coming to similar conclusions. Take for example last fall's "Low-salt Diet Increases Insulin Resistance in Healthy Subjects". Insulin resistance is a big risk factor for diabetes and diabetes is a big risk factor for CVD.

So what to make of the efforts of the low-salt advocates to subject all Americans to what amounts to a mass uncontrolled experiment without their consent (informed or otherwise) and without the oversight of an ethics review board that might otherwise be charged with safeguarding the public? And what to make of the absence of the Precautionary Principle typically unfurled before our self-styled public health advocates march us out to face nature's cannons? Apparently "caring" means never having to say you're sorry.

But should Uncle Sam have to say he's sorry and pay for the tens of thousands of unnecessary deaths that would occur annually if the findings of these new studies are indeed predictive of what would happen if low-salt diets were forced upon (nudged upon) the entire unwitting populace? If the asbestos story is any indication then the food and beverage industry needs to prepare to be thrown under the bus at the earliest sign that the central low-salt diet plan has not survived first contact with the Law of Unintended Consequences. Because that's exactly what the government that stockpiled asbestos and specified its use did to those whom it directed to stockpile and produce it when asbestos hit the fan nearly 40 years ago.

More on that, plus an old depo of a government witness or two, later.

Getting at the Truth: Improving Systematic Reviews

Systematic reviews and meta-analyses are considered to be perhaps the best evidence about treatments/exposures and outcomes from which causal inferences can be drawn. The problem is that they're susceptible to a variety of biases. See e.g. "Bias Due to Changes in Specified Outcomes During the Systematic Review Process".

In the litigation context we increasingly see a systematic review launched by an expert-to-be before the first suit is even filed.  They will thus have already gone through the studies; used their "judgment" to select which ones ought to be considered (and which ought not); determined how much weight to give each; and established a protocol for conducting the review which, surprise surprise, demonstrates a causal link supporting the lawsuits. The strong suspicion is that selection bias (a form of the Texas Sharpshooter effect) is at work (and that, of course, is being charitable). And the biggest problem is that uncovering selection bias is notoriously difficult.

Now there's a move afoot to bring transparency to systematic reviews by requiring that plans, methods and protocols be registered before the systematic review is begun. See "Best Practice in Systematic Reviews: The Importance of Protocols and Registration"; "New Initiative to Make Systematic Review Protocols More Transparent"  and "Open Medicine Endorses PROSPERO". Here's the press release from the Cochrane Collaboration and here's a link to PROSPERO.

Until we get courts to make experts disclose their methods and justifications for selecting, weighing and interpreting data before they develop their opinions articles like the following may come in handy: "How Can We Improve the Interpretation of Systematic Reviews?"

Do Transient Changes in the Levels of PM2.5 Cause Ischemic Stroke?

Recent attempts to lay pretty much any cardiovascular or pulmonary malady at the feet of fine particulate matter got a boost with JAMA's publication this month of the commentary "Air Pollution as an Emerging Global Risk Factor for Stroke". But what happens when you actually monitor the air for PM2.5 and then compare the levels found to the rate of ischemic stroke? If there's an association you'd think rising levels of PM2.5 would be followed by an increase in ischemic stroke. "Fine Particulate Air Pollution (PM2.5) and the Risk of Acute Ischemic Stroke" found no support for such an association (though it should be noted that an increase was detected among one subgroup: diabetics)

More Evidence That Electric Currents and Electromagnetic Fields Are Not Associated With Childhood Leukemia

See "Exposure to Electrical Contact Currents and the Risk of Childhood Leukemia" in this month's Radiation Research journal. "In this California population (245 cases and 269 controls from the Northern California Childhood Leukemia Study), there was no evidence of an association between childhood leukemia and exposure to contact currents or magnetic fields.."

Population Mixing and Childhood Leukemia: Is a Mosquito to Blame?

When Pat Buffler co-authors something I take notice. Of the many epidemiologists I've run across her reputation for sound science is perhaps the most unassailable. In "Unusual Space-Time Patterning of the Fallon, Nevada Leukemia Cluster: Evidence of an Infectious Etiology", Francis, Selvin, Yang, Buffler and Wiemels examine the Fallon childhood leukemia cluster and find that most of the cases arose during the early to mid summer. Furthermore, a parallel increase in childhood leukemia was found to exist among the children of the military personnel newly introduced to the area. Together these findings, along with information about the habits of the local mosquitos, suggest an infectious cause for the childhood cancer cases that followed.

Could a single event cause cancer weeks later? Doesn't it take years for cancer to unfold? Make chromothripsis your word of the day.

Finally, you plaintiff's lawyers working on childhood leukemia cases, instead of using experts who can only get their opinions published in low impact journals of which they're among the editors (see e.g. Green v. George's Farms, et al), consider instead the impact of, bah, you figure it out. It's obvious at this point.

 

Hammar Time

The Federal MDL's Magistrate Strawbridge has held that Dr. Hammar's opinion "that 'every occupational and bystander exposure to asbestos above background was a substantial contributing factor in causing [plaintiff's] mesothelioma' is sufficiently reliable to meet the admissibility standard of Rule 702 ..." See In Re Asbestos Products Liability Litigation Relating to: Anderson v. Saberhagen Holdings, Inc.

Because Dr. Hammar testified that his theory that every fiber is causative "could be (though it hasn't been) proven from carcinogenic theory (whatever that is) to a reasonable degree of medical certainty" and because "any of the occupational or bystander exposures could have been (though again they haven't been) sufficient to cause [plaintiff's] mesothelioma ..." his testimony ought not be excluded.

So Dr. Hammar gets to testify about legal duty (substantial factor causation) and base his opinions on unverified hypotheses. Oh, and in another case, plaintiffs get to recover money from the trust funds and then deny that their application forms don't establish substantial factor causation. See In Re Asbestos Products Liability Litigation Relating to: Taylor v. Lucent Technologies NC.

Wow. Just wow.

A Big Step Forward in Tracing Foodborne Illness Back to Its Source

Where pulsed-field gel electrophoresis failed to distinguish a strain of Salmonella enterica responsible for a major salmonellosis 2009-2010 outbreak from a strain isolated in a prior separate outbreak so-called next-gen DNA sequencing (NGS) has allowed researchers to pinpoint the source of the recent outbreak to a specific spice used on salami.  See Identification of a Salmonellosis Outbreak by Means of Molecular Sequencing.  Note however that traditional epidemiologic traceback techniques had lead investigators to the source of the outbreak.  NGS confirmed the outbreak and further demonstrated that endemic contamination of food facilities by pathogens can lead to the production of new strains thereby making more precise sequencing "essential to the traceback of bacterial pathogens as they emerge in the food supply."

Does Timing of Hormone Therapy Affect Breast Cancer Risk?

Initiating hormone replacement therapy before or shortly after the onset of menopause is associated with a greater risk of developing breast cancer compared to those women who started hormone therapy after living with menopause for a few years or more. That's according to (ungated) "Breast Cancer Risk in Relation to Interval Between Menopause and Starting Hormone Therapy".

Other findings include: (1) further confirmation of the observation that hormone therapy increases both false positives and false negatives on mammography; (2) obesity increases the risk of breast cancer in women not on hormone therapy though it has no effect on the risk of those undergoing hormone replacement therapy; (3) contrary to other observations, in this Million Woman cohort risks did not revert to background following cessation of therapy indicating that hormone therapy may not merely "fuel the fire" but may in some cases start it as well; and, (4) the highest risk of breast cancer was seen in women currently on estrogen-progestin who had begun the therapy less than 5 years after the onset of menopause.

Infants on Antibiotics Experience a Big Increase in Their Risk of Developing Asthma

Even if neither parent had asthma, introducing antibiotics to a child less than six months of age produced a big increase in the risk that he or she would develop asthma five and a half years later. Do antibiotics prevent the establishment of bacteria in our gut that help us recognize good from bad in the outside world and modulate our immune system's response or do antibiotics nuke our gut microflora dysregulating our immune systems? Read all about it (free online) in "Antibiotic Exposure by 6 Months and Asthma and Allergy at 6 Years: Findings in a Cohort of 1,401 US Children" in the American Journal of Epidemiology.

Population Mixing, Childhood Leukemia, Viruses and Vaccines

There's good epidemiological evidence that population mixing is responsible for several clusters of childhood leukemia (acute lymphocytic leukemia, or ALL). Some have hypothesized that viruses are to blame but there hasn't been much evidence to support that hypothesis; at least not until now.

In the current Journal of Pediatrics you'll find "Associations Between Vaccination and Childhood Cancers in Texas Regions" which compares the risk of ALL to vaccination rates in different public health regions. With all the caveats that must go along with hypotheses generated by statistical analysis it is nevertheless quite intriguing to see that children vaccinated against a wide range of viruses had a large and consistent reduction in their risk of ALL; so much so that it leading the researchers to conclude that "[s]ome common childhood vaccines appear to be protective against ALL at the population level."

Be sure to also note that 4 doses of diptheria-tetanus-pertussis, 3 doses of polio, 1 dose of measles-mumps-rubella, 3 doses of H. influenza, type B, 3 doses of hepatitis B and one dose of Varicella, the 4-3-1-3-3 vaccine regimen claimed by some anti-vaccine activists to be capable of "overloading young immune systems" and thereby (some-unstated-how) causing autism, produced a 38% decrease in the risk of a child developing leukemia.

Hopefully the anti-vaccine crowd is paying attention. The list of harms to children for which they may be made to answer is apparently growing.

Does Recall Bias Explain Past Associations Between Pesticides and Parkinson's

How do people's memories of pesticide exposures correlate with industrial hygiene estimates of those exposures? Not so well. In fact it's pretty clear that a lot of people with Parkinson's assume that chemicals caused their illness and so are primed to remember past high exposures that had not in fact occurred. For a well done paper showing no association between pesticides and Parkinson's plus a great discussion of recall bias see: "Pesticide Exposure and Risk of Parkinson's Disease - A Population-Based Case-Control Study Evaluating the Potential for Recall Bias".

Those looking for the real cause of the increase in risk of Parkinson's among those involved in farming should pay attention to the endotoxin discussion. I'll check the studies that show endotoxin may protect against lung cancer in cotton textile workers to see if there's any hint of a Parkinson's excess and report back.

Do Antibiotics Administered Early in Life Cause Inflammatory Bowel Disease and Crohn's?

There's a growing body of evidence pointing to early disruption of gut microbiota via antibiotics as a cause of inflammatory bowel disease and Crohn's disease later in life. For the newest see "Association Between the Use of Antibiotics in the First Year of Life and Pediatric Inflammatory Bowel Disease" in this month's American Journal of Gastroenterology.

Another Chinese Study of MDS Risk Factors

So what are the myelodysplastic syndrome (MDS) risk factors du jour? Treatment for tuberculosis, benzene, "new building and renovation", pesticides, hair dyes, traditional Chinese medicines (for one subtype of MDS), gasoline and herbicides. Education was somehow protective. See "Case-Control Study of Risk Factors of Myelodysplastic Syndromes According to World Health Organization Classification in a Chinese Population" published in the Feb. 2011 American Journal of Hematology.

Breast Implants and Cancer

I thought it was no biggie when the FDA sent out an email late Wednesday morning saying that an extraordinarily rare malignancy, anaplastic large cell lymphoma (ACLC), had been associated with breast implants. A variety of implants, mainly orthopedic devices, have long been associated with certain rare cancers. Since the site of the cancer tends to coincide with the site of complicating surgical infections it has been thought that an infectious agent was responsible. See e.g. "Soft Tissue Anaplastic Large T-Cell Lymphoma Associated With a Metallic Orthopedic Implant: Case Report and Review of the Current Literature".

A quick review of PubMed showed that concern over ACLC and breast implants had been around for years. See e.g. "Anaplastic Large-Cell Lymphoma in Women With Breast Implants" (free in JAMA) published in 2008. So I went looking for something else to post on. Then, on tonight's 10 o'clock news here in Houston, one of the local stations led off with a story about the late John O'Quinn's litigation against Dow Corning and his claims that silicone implants caused autoimmune disorders and cancer. They made it sound as though O'Quinn had had somehow been vindicated by today's FDA press release. Then they went out and found some sympathetic woman who had recently had a radical mastectomy followed by breast reconstruction and asked her what she thought about the "new report on breast implants and cancer". To her everlasting credit she said she was happy with her decision and was confident that she'd made the right one.

ACLC is not breast cancer and the odds of getting it, assuming the association is confirmed (and there is indeed an awful lot of evidence showing that in the areas around implants whether silicone or metal where infections can set in, cancers can sometimes follow) are about 1 in 900,000. The odds by the way of drowning in your bathtub are significantly higher - somewhere around 1 in 660,000.

The media could have focused on the story of the mounting evidence for a link between pathogens and cancer. Instead they seem to have resorted to a long since discredited narrative about breast implants. It's too bad because the real story is the story of our generation.

A Disingenuous Take On The Vaccine-Autism Fraud

The British Medical Journal has just published an editorial titled "Assuring Research Integrity in the Wake of Wakefield" that addresses what has finally been revealed to have been an elaborate fraud concocted by a scientist and some personal injury lawyers in an effort to launch a mass tort. Unfortunately, rather than addressing the real problem (which is that the majority of the published peer-reviewed papers purporting to find an association between some drug or exposure or gene and a disease are probably false) the authors of the editorial reference a handful of ethical lapses spaced about twenty years apart and ask "[h]ow could this happen again?"; implying rather obviously that scientific fraud is almost as rare as Piltdown Man but nonetheless something about which the academy ought to be vigilant lest the public lose faith in "science".

They conclude with "We must transcend traditional hierarchies and authority gradients to empower everyone in the research enterprise ... to raise questions and "stop the line". I've no idea what the first part means though it sounds suspiciously like something out of "Transgressing the Boundaries: Towards a Transformative Hermeneutics of Quantum Gravity". The latter part on the other hand, quite inadvertently I assume, manages to expose the real problem with today's "research enterprise". It refers to the ability of factory workers on an automobile assembly line to halt the process when they detect a problem rather than having to wait until a supervisor calls for a stop. Focus then on the idea that many "researchers" aren't involved in the process of discovery or even design. Rather their part is played down on the assembly line of the Science factory - manufacturing the same sort of science; shift after shift, day after day, year after year. Their job is to identify anything that might throw a wrench in the works or cause the product to be defective and thus rejected by the customer, typically the government, industry or an NGO, to repair or engineer around it and to keep the line running.

The authors' concern then is with the process and not the product. But of course, if you've been paying attention, you know by now that the product is the real problem. In studies like Wakefield's, in which statistics are trotted out to test hypotheses, the "science" is probably wrong even if the researcher isn't consciously cooking the books in order to gin up a mass tort. Read and re-read "Odds Are, It's Wrong" from ScienceNews. Let the following quote sink in: "There is increasing concern that in modern research, false findings may be the majority or even the vast majority of published research claims."

Whether or not the product of the Science factory is worth its price or indeed worth anything at all, it keeps on coming in an ever increasing torrent.  Take for example genome-wide association studies (GWAS) - one of the most notorious examples of  too often useless "research" produced assembly line style. (Note there are efforts to improve it. See e.g. "A Knowledge-Based Weighting Framework to Boost the Power of Genome-Wide Association Studies") A quick search of PubMed reveals that 45 new genome wide association studies have rolled off the line in just the last week. That's great news if you're in the business of selling SNP chips to research universities and a sign of a boom (or bubble) in the fortune of researchers. And maybe it's even good for the economy - being after all a form of digging holes and filling them back in. But where does it end?

Ponder the following from "The Future of the Research University" written in 1997: "We need to think seriously, within the community of research universities, about whether we are producing too many Ph.Ds. This is a controversial question, with different answers in different scholarly disciplines, but the general conclusion seems inescapable: The mathematidcs of exponential growth - each professor producing numerous Ph.Ds who become professors who produce numerous Ph.Ds, etc. - has caught up with us."

With that exponential growth in Ph.Ds desperate for something to research and something to publish it's no wonder that so many turn to statistical tools which, when rigorously and repeatedly applied to any mound of data, will inevitably produce a publishable statistically significant, though often false, result. And despite concern that the exponential growth in the number of Ph.Ds "has caught up with us" there's no sign that the factory is cutting back on the number of shifts. Instead, more factories are being built. Indeed what got me thinking about this was a huge billboard on I-45 announcing that what was once the humble but excellent Teachers' College is now itself a Research University! Sure enough, a stroll through their website reveals that they've bought a great pile of extremely expensive analytical equipment and will soon be adding to the mountain of Science being manufactured. Everybody it sometimes seems is getting in on the "research enterprise".

The point then is that research has become an industry; and an enormous one at that. Best then to stay as skeptical of Big Research and Big Research Publication (note the circle the wagons approach of the Lancet's editorial board when they first got whiff of the fraud) as you are of Big Corporation. And best then as well to train your attention on the product of factory Science even if the process by which it was made is sound. Remember, it's generally not enough to say simply that it was made According to Plan.

Dry Cleaning Workers: End-Stage Renal Disease, Esophageal, Lung and Tongue Cancer

Here's a small study (1704 workers) of people exposed to "perc" (perchloroethylene a/k/a tetrachloroethylene): "Mortality and End-Stage Renal Disease Incidence Among Dry Cleaning Workers". Small increases in the overall risk of dying from cancer of any kind and from esophageal, lung and tongue cancers in particular were identified. Most significant was a near doubling of the risk of developing end-stage renal disease. As risk appeared to rise with increasing duration of employment (a proxy for exposure) at least for some of the cancers it was concluded that perc, rather than lifestyle or socioeconomic factors, was responsible and that it's therefore likely a human carcinogen.

Yet just last October a study of 10,389 Swedish dry-cleaning and laundry workers (the largest yet of perc-exposed workers) found no increase in cancer risk either overall or for those sorts of cancers thought possibly related to perc exposure. The article, which is free, has a great discussion of the perc controversy, the various positions taken by IARC, the NTP, the ACGIH and others, and a brief discussion about why the healthy worker effect might not be an issue in occupational cancer inquiries. It was published in the International Archives of Occupational and Environmental Health as "Cancer Morbidity in Swedish Dry-Cleaners and Laundry Workers History: Historically Prospective Cohort Study".

El Dorado's Lost City of Uranium and Good Health?

In epidemiology, whenever lower incidence and lower rates of mortality from cancer occurs in a population commonly assumed to be at risk cognitive dissonance is always lurking. What generally happens is that the good news is shrugged off as "the healthy worker effect" and epidemiologists resolve to re-sift the data in order "to get the right answers". The "right answers" of course are often only those that support our preconceptions.

That means there aren't many people willing to even consider the possibility that working in a chemical plant or going off to war or spending a career mining / processing uranium while being exposed to low levels of gamma radiation might actually confer a health benefit. Nevertheless, the so-called healthy worker effect (which is called the healthy warrior effect for those who served in the armed forces) appears so commonly and across so many trades that you have to wonder if something besides simply screening employees for good health is at work. If you're interested here are three studies worth pondering.

In this month's journal of Occupational Medicine you'll find an excellent discussion in "The Healthy Worker Effect in US Chemical Industry Workers". The study was of thousands of Dow Chemical employees - three million years worth of employment combined. The overall risk of death from any cause was lower than expected as was the risk of dying from nine types of cancer thought to be related to smoking. Nevertheless the authors, an epidemiological team working for the Dow Chemical Company, concluded that the findings were likely due to the healthy worker effect though with a caveat. Some have suggested that the healthy worker effect arises because employers dismiss employees with health problems. However, in this study the health outcomes of those employed for a decade or more were the same as those who didn't last very long with the company. The finding thus suggests that the presumed healthy worker effect was generated as each employee was considered for employment such that workers destined to get cancer decades later somehow were never hired in the first place making it in fact a "healthy hire effect".

For another example see "Psychiatric Diagnoses in Historic and Contemporary Military Cohorts: Combat Deployment and the Healthy Warrior Effect". Despite some claims in the media that might make one assume otherwise those who serve in the military see lower than expected numbers of ailments including psychological ones. Here the suggestion is that those prone to psychological illness  are screened out as what risk there was of developing psychological issues after combat was concentrated in those with preexisting mental health issues.

Finally, published last month in the journal Radiation Research, there's "Mortality (1950 - 1999) and Cancer Incidence (1969 - 1999) in the Cohort of Eldorado Uranium Workers". With the exception of lung cancer incidence and mortality which demonstrated a small increase in risk, the Eldorado uranium miners managed to have significantly lower risks of dying from any cause, lower risks of dying from all cancers combined (lung cancer included) and a lower risk of developing any type of cancer cumulatively (lung cancer again included).

Now be honest. If someone had asked you yesterday whether you'd pick (A) uranium workers, or (B) the average Canadian male, as the group likely to have a much larger risk of getting cancer and of dying of cancer which would you have chosen?

So what's at work here? Is it simply that those prone to developing cancer in the distant future are somehow weeded out and never hired in the first place? Or does it perhaps have something to do with the nature of blue collar employment over the last 50 years? To me it all looks a lot like the compliance effect - the phenomenon whereby e.g. those who lead very ordered and structured lives and who thus always take their placebo at the appointed time manage not only to do better than those less disciplined and also on a placebo but oftentimes better even than those less disciplined but at least irregularly taking real medication. So I'll go out on a limb and guess that the claims of toxic tort plaintiffs notwithstanding, large employers engaged in manufacturing not only didn't shorten the lives of their workers they lengthened them by imposing the very order and rigidity about which so many bitterly complain in their depositions.

Human Papillomavirus Expected to Cause More Than a Million Cancer Cases

Human papillomavirus, or HPV, is well established as a cause of cervical cancer. Now it's thought that the infection causes even more cases of head and neck as well as anogenital cancers. For the newest on HPV and cancer see:  "Impact of HPV on Oropharyngeal Cancer". Obviously prevention is important and new vaccines appear to be safe and effective ("Efficacy and Safety of Prophylactic Vaccines against Cervical HPV Infection and Diseases among Women: A Systematic Review and Meta-Analysis") but what to do about the millions already infected? For some thinking along those lines see: "Human Papillomaviruses As Therapeutic Targets in Human Cancer". 

Also of interest is a new discussion of the evidence for the proposition that male circumcision protects women from cervical cancer in "Male Circumcision and HPV Transmission to Female Partners". Combine such rediscoveries with the "Oops, sorry" reaction of those, including the WHO, who had recommended that infants be breast fed exclusively for six months (and who now must eat humble pie in light of studies showing that such advice, if followed, likely leads to food allergies or worse) and perhaps it's time for a new approach to activist-led health initiatives that seek to upend traditional practices. Since activists can't be made to pay for the injuries they cause (mass tort litigation generally being a search for deep pockets rather than truth) maybe we should at least require that any demand for an extraordinary change in traditional practices be backed up by extraordinary evidence.

Of Flame Retardants, Autism and Skepticism

Last October some scientists got together in San Antonio to discuss the potential hazards of flame retardants. They wound up signing the "San Antonio Statement on Brominated and Chlorinated Flame Retardants" co-authored by Ake Bergman. Their claim is that the flame retardants are bio-persistent, toxic (especially when burned), cause neurological development in children and, because of their use in electronics including housings, being effectively dumped in third world countries where the products are recycled.

Bergman was interviewed for The Researcher's Perspective and you can read or listen to the interview at EHP in "The San Antonio Statement, with Ake Bergman". Bergman is quoted as saying of flame retardants "they are acting in a similar way than [sic] the other chlorinated compounds are, which is leading to a number of effects - for example, cancer risks; endocrine-disrupting properties of the chemicals, we have reproductive effects of the chemicals; and not the least, the neurodevelopmental effects that they cause, and for the neurodevelopmental we are talking about young children, the newborns, being affected." He goes on to say that he hopes that five years hence such flame retardants will all be banned and he says that "it's ridiculous to learn that you have nursing pillows with flame retardants..." Pity the poor maker of nursing pillows. A dropped cigarette or a pillow set too close to a space heater and woe be to the manufacturer who made it of cotton and didn't soak it in flame retardants.

Anyway, what should we make of the claim of neurodevelopmental effects? Well, as fate, or luck, would have it Dr Bergman has just co-authored a newly published study titled "Polybrominated Diphenyl Ethers in Relation to Autism and Developmental Delay: A Case-Control Study". The data from the study show that there is no relationship between PBDEs (the flame retardants at issue) and autism. In fact, the researchers managed to find that children suffering from neurological development delay were the ones with the lowest exposures to flame retardants.

So what did the researchers have to say about these findings? See pages 16 - 20 of the paper. Rather than simply reporting "we found no association between neurological development and exposure to flame retardants" the authors spend five pages saying why their exposure data is probably wrong and why even more studies of flame retardants and neurological development are needed.

Wouldn't it be nice, just once, if a scientist found an association between a chemical and some hot button disease and she spent her entire Discussion and Conclusion pointing out the reasons not to panic and not to jump to conclusions? That it doesn't happen when associations are found but does happen when the null hypothesis is confirmed says it all.

Nulliparous Plaintiffs, Fault and Causation

It has been known for a couple of decades now that women who never have children (i.e. women who are nulliparous) and women who do have children but not until they are 30 or older suffer a striking increase in their risk of developing breast cancer. The evidence for the association between never giving birth or delaying having a child continues to accumulate and now it appears that the increased risk is focused on hornmone receptor-positive breast cancers. See "Associations of Breast Cancer Risk Factors With Tumor Subtypes: A Pooled Analysis From the Breast Cancer Association Consortium Studies" in the current issue of the Journal of the National Cancer Institute. So let's say you've got a nulliparous plaintiff alleging that your drug or device or chemical caused or accelerated her hormone receptor-positive breast cancer; how do you handle her status?

The first problem a defendant faces in such a case is the risk of inadvertently wandering into the minefield called "blaming the victim". The plaintiff has either freely made a choice or has tragically been unable to have a child. Either way the jury will react strongly and negatively to any discussion about parity status and causation that makes even the slightest trespass into the issue of fault. Keep the discussion limited to risk factors and their relative potency. But that leads to another problem.

In some of the jurisdictions in which I practice plaintiff's counsel will successfully argue to the trial court that only evidence about about the actual cause of plaintiff's injury is admissible. In other words, unless my expert is prepared to say e.g. that "to a reasonable degree of medical probability plaintiff's breast cancer was caused by her not having children when she was young" testimony about "mere risks" is irrelevant and so inadmissible. The practical effect of such a ruling is that only junk science is admissible on the issue of the actual cause of plaintiff's cancer since my experts tend to be modest about the claims science can make regarding the cause of any individual's cancer. We're stuck then trying to prove a negative, showing we acted reasonably and preserving error.

In this age in which much that was certain (e.g. that we've conquered infectious diseases) is proving not to be so it's time I think for courts to recognize not only that the reasonableness of actions can fairly and effectively be judged according to the risks they conferred but also that causation is in many cases most precisely weighed when competing risks are allowed to be compared against one another.

Finally, and hopefully still on topic, for more evidence of the complexity of causation see "Does Pregnancy Provide Vaccine-Like Protection Against Rheumatoid Arthritis?" Why would pregnancy protect against auto-immune disorders and what's the connection with breast cancer? There are a variety of hypotheses offered but so far no one knows.

 

Wealthy and Healthy

Wealth strongly and consistently correlates with good health and an overall reduced mortality risk. A new paper summarizing past research and presenting new data which confirms the link has just been published in the American Journal of Epidemiology. It's titled "Long-Term Effects of Wealth on Mortality and Self-rated Health Status". The study focused on self-reported perceptions of health status and the results mirrored those of the objective measure for mortality: socioeconomic status is a good predictor of health status.

Why do the least wealthy (aka the poor) tend to be the least healthy? Those pushing the cause of so-called environmental justice claim that the poor, who unsurprisingly live in the cheapest and thus least pristine areas, are exposed to toxic chemicals, electro-magnetic fields and ionizing radiation at levels far higher than the rich (who tend not to build their mansions next to refineries) and that such exposures are to blame. Others, the "real food" activists, claim that the poor live in a junk food environment and, the W.I.C. program notwithstanding, have not the means to come by nutritious food. Still others claim that the poor suffer from inadequate health care. There is though another reason the poor might be so afflicted. It was best stated by a now deceased safety man who spent his career with one of the oil companies in Port Arthur, TX as follows: "Poor folks got poor ways."

We were at the deposition of a retired refinery worker who was suffering from leukemia and who was giving a deposition before he passed away to be used by his wife in a subsequent gross negligence case against his and the safety man's employer. The deponent was asked by plaintiff's counsel, "If you'd have known about the dangers these chemicals like benzene posed would you have come to work for xxxx Oil Company?" The man, obviously prep'd for the question answered "No way. I'd 'a stayed in the piney woods loggin' like my daddy done. It might not 'a paid as much but I wouldn't 'a got this cancer". The safety man leaned over and said "Yeah, and he'd have died of cirrhosis at 48 just like his daddy done."

That old safety man went on to explain that the men who came out of the woods, and the cane fields and off the pogy boats to work hourly at those refineries often signed their job applications with an "X" but they wound up solidly in the middle class and their children or grandchildren made it to college and beyond. "Some couldn't even write their name when they got here and nobody ever went back to bein' a gawddamned logger from the xxxx Oil Company."

The epidemiological studies of those refinery work forces have repeatedly found that overall the men lived significantly longer, were significantly healthier and had a lower risk of all cancers combined than similar men in the general U.S. population - despite exposures to asbestos, benzene, butadiene and the like. Compared to loggers, and boat crews and farm laborers the average refinery worker bought himself six more years of healthy life just by going to work for better pay in an environment where middle class values were the norm.   

More Evidence That Exposure to Poultry Viruses and Bacteria Causes Cancer in Humans

Significant increases in mortality risk for some forms of leukemia has again been identified in a cohort of poultry workers. See "Update of Cancer and Non-Cancer Mortality in the Missouri Poultry Cohort". Given the profound changes in the demographics of the poultry industry in recent years it would be interesting to see if population mixing might have been responsible for some or all of the increased risk.

Peter Infante Isn't Happy About IARC's 2009 Benzene Evaluation

A member of NIOSH's 1975 Benzene Task Force and long-time plaintiff expert, Infante has written essentially an editorial comment about IARC's determination that there is only limited evidence for benzene causing blood cancers other than AML (or ANLL or whatever it's called this week). It's published online first in the American Journal of Industrial Medicine as "The IARC October 2009 Evaluation of Benzene Carcinogenicity was Incomplete and Needs to be Reconsidered."

Infante was on plaintiff's team in the case that resulted in the biggest check ever written to a single plaintiff in a benzene case - Mason v. Texaco.

What To Do About Too Many Calories for "Sedentary Young Children"?

The Center for Science in the Public Interest (CPSI) has sued McDonald's over their Happy Meals. The complaint argues that Happy Meals are "unhealthy" and cause obesity, that its marketing is "unfair" because it makes six year old lead plaintiff Maya pester her mother with her "clamor for Happy Meals" and that McDonald's seeks by way of its Happy Meals, like James Dean before it, "to subvert parental authority".

There is much in the complaint to blog about. Far too much actually given that I've got a brief due in the Texas Supreme Court by Friday. For now though consider the claim that Happy Meals provide too many calories for the typical, which is to say sedentary, child. When and how did the typical child get to be sedentary and so at risk of obesity? I'd argue that it has everything to do with turning schools into warehouses for children.

Want some evidence that even moderate exercise protects children from overeating or too much TV/video-gaming? See "Health Status and Behavior Among Middle-School Children in a Midwest Community: What are the Underpinnings of Childhood Obesity?" See also "The Fat-Mass and Obesity-Associated (FTO) Gene, Physical Activity, and Risk of Incident Cardiovascular Events in White Women". Apparently the "fat gene" only causes problems when combined with inactivity.

All in all it looks like the solution to childhood obesity is more about revving up the body's engine than starving it of fuel.

Yellow Water: No Need to Have Panicked?

Hexavalent chromium (Cr (VI)) turns well water yellow. Disgusting, sure, but is it harmful? Well, when inhaled Cr (VI) can cause lung cancer in humans; and that's all it took to launch a lawsuit on behalf of the citizens of Hinkley, CA whose groundwater had been contaminated with Cr (VI). The case settled for $333 million and spawned the movie Erin Brockovich.

So what was the cancer burden from Cr (VI) in Hinkley? According to a new study reported on in the LATimes (Fewer Cancers Found in Hinkley Than Expected)  the answer is none whatsoever. In fact, while 224 cancers would have been expected in an unexposed population of similar size and demographic make up, only 196 cancers were identified in Hinkley - more than a 12% decrease in cancer.

There's not much new in the literature on Cr (VI) and the claim that hexavalent chromium in water causes cancer is at best controversial. The newest paper supporting a link, "A Quantitative Assessment of the Carcinogenicity of Hexavalent Chromium by the Oral Route and its Relevance to Human Exposure" found that long suffering lab mice exposed to many times the levels in Hinkley developed cancer of the small intestine (not among the complaints chronicled by the panicked residents in the LATimes' report). The finding was said to provide further support for the claim that Cr (VI) is "likely to be carcinogenic to humans". Nevertheless, epidemiologists have yet to identify a clear case in which it actually has been carcinogenic to humans.

Risk Factors For Pancreatic Cancer

Smoking and H. pylori are at the top of the list. Further down you'll find much weaker risks like alcohol, diabetes, metabolic syndrome and having a non-O blood type. Here's the newest review: "Epidemiology of Pancreatic Cancer: An Update".

Pesticides and Prostate Cancer: A Two-fer

Epi + Biomarkers = "An Update of Cancer Incidence in the Agricultural Health Study" + "Pesticide Use Modifies the Association Between Genetic Variants on Chromosome 8q24 and Prostate Cancer".

"A Global Toxic Emergency"?

The Story of Stuff people have now released "The Story of Electronics". In it they proclaim "a global toxic emergency" declaring that Silicon Valley is "one of the most poisoned communities in the U.S." and repeat an old canard about the alleged toxic perils of working in high tech manufacturing in general and clean rooms in particular. As is typical of such pieces even the most modern facilities are portrayed as little better than the Triangle Shirtwaist Factory.

Coinciding with the video's release comes the publication of its antidote. In "Cancer Mortality Among US Workers Employed in Semiconductor Wafer Fabrication" the health outcomes of 100,081 semiconductor workers employed between 1968 and 2001 were examined and analyzed. Compared to people in the general population of similar age, gender and race their overall risk of death from all causes combined was cut in half and their risk of dying of any sort of cancer was just three quarters that of other people. As for cancers of the blood referenced in "The Story of Electronics", there's nothing to worry about in that regard either.

The paper concludes "Work in the US semiconductor industry, including semiconductor wafer fabrication in cleanrooms, was not associated with increased cancer mortality overall or mortality from any specific form of cancer."

The exposure assessment for these same workers was also just published and can be found in "Exposure Assessment Among US Workers Employed in Semiconductor Wafer Fabrication".

Another Paper Reports a Strong Association Between Chrysotile and Lung Cancer

Interestingly, there's a dramatic difference in risk between those with heavy and those with moderate exposure to chrysotile asbestos and, in fact, the small increase in risk to those with moderate exposure did not reach statistical significance. See "Lung Cancer Mortality From Exposure to Chrysotile Asbestos and Smoking: A Case-Control Study Within a Cohort in China".

Cell Phones and Brain Cancer: What Was The New York Times Thinking?

Recently the NYTimes published "Should You Be Snuggling With Your Cellphone?" in which it reported that the question of whether cell phones pose a risk of brain cancer is far from settled. Indeed, largely ignoring the overwhelming evidence that electromagnetic radiation does not increase the risk of brain cancer the article references an unidentified study showing an increased rate of brain cancer in the presumably cellphonophilic 20-29 year old age group, "400 scientific papers" that support the theory that radio waves cause "damaged brain DNA", and concludes with this quote by epidemiologist and author of the newly published "Disconnect" Devra Davis: "I do think I'm looking at an epidemic in slow motion". 

Google serves up a link to Environmental Health Trust which has front and center handy links to a "press kit", and variety of write-ups including (1) that the best evidence does support a causal link; (2) a warning to women to keep cell phones away from their chests as the radio waves "seep directly into soft fatty tissue" and may cause breast cancer; (3) heavy cell phone use decreases sperm count by 50%; (4) that we should be horrified by the sight of young children using iPhones as they are frying their young brains; (5) a conspiracy by industry and lobbyists to obfuscate the facts and prevent urgently needed anti-cell phone legislation; and, (6) the inevitable lawsuit by someone with an unidentified form of brain cancer who claims cell phones are as addictive as cigarettes and just as deadly - the evidence that cell phones caused his cancer seems to be limited to the fact that he was athletic, a non-drinker / non-smoker who led "an over-the-top healthy lifestyle".

Well, increasing risk of brain cancer in young adults, "damaged brain DNA", a corporate conspiracy and a plaintiff who talked on his phone four hours per day 365 days per year, how is this NOT the start of a massive mass tort? Here's how.

Let's take that troubling trend of increasing brain cancer in 20 - 29 year olds. Open up this month's journal of Neuro-Oncology and you'll find "Brain Cancer Incidence Trends in Relation to Cellular Telephone Use in the United States". Yes, there's small increase in incidence for males but the increase began "before cell phone use was highly prevalent". Yes there was an increase for women in that age group too but it was limited to frontal lobe cancers and since I've never seen a woman use her cell phone by holding it to her forehead I have to wonder if the absence of any increased risk in brain cancers near the ears isn't the most important finding of this huge study. And in fact its authors conclude "these incidence data do not provide support to the view that cellular phone use causes brain cancer". See also: "Mobil Phone Base Stations and Early Childhood Cancers: Case-Control Study" which found "no association between risk of early childhood cancers and estimates of the mother's exposure to mobile phone base stations during pregnancy" and, of course, "Brain Tumour Risk in Relation to Mobile Telephone Use: Results of the INTERPHONE International Case-Control Study" which showed that most cell phone users were if anything at a decreased risk of cancer.

Damaged brain DNA (whatever that is)? "Analysis of Proteome Response to the Mobile Phone Radiation in Two Types of Human Primary Endothelial Cells".

And the claim of industry bias or outright conspiracy to silence the truth or co-opt scientists? Try "Studies of Mobile Phone Use and Brain Tumor Risk Are Independent of Industry Influence".

Finally, brain cancer is always tragic, whether it strikes down a U.S. Senator or my next door neighbor at age 48; and their drinking, smoking and exercising status makes no difference as none are risk factors, positive or negative.

To this day the cause or causes of brain cancer remain unknown. All you can do is drink your coffee or tea and hope that one of life's deadly bolts out of the blue doesn't strike you.

Childhood Leukemia and Population Mixing: The Evidence Mounts

Despite numerous attempts to blame anything and everything associated with a deep pocket for a cluster of childhood leukemia cases in Fallon, NM,  the strongest evidence by far supports the hypothesis that an infection, spread by the sudden mixing of two or more populations, was responsible. Now there's evidence that childhood leukemias previously blamed on nearby nuclear reactors are in fact strongly associated with population mixing. See "Childhood Leukaemia, Nuclear Sites, and Population Mixing".

That introducing lots of new workers to a formerly rural environment to build a military base or nuclear power plant often is accompanied by an increase in childhood leukemia cases is now confirmed by more than a dozen studies. That often unnoticed infections cause cancer is also well established. So, what duty, if any, is owed to the local children and to the children of its own employees when a business sets up say a slaughterhouse in rural North Carolina and staffs it with workers from a long way away? And if there is a duty, how is it to be fulfilled? To warn? Of what? To monitor? For what? To protect? With what?

More Evidence That Exposure to Diesel Exhaust is Causally Associated With Lung Cancer

The increased risk is smallish, with an odds ratio of 1.31, but the finding was consistent, showed a strong dose-response relationship and statistically significant (95% CI of 1.19 - 1.43). Read about it in "Exposure to Diesel Motor Exhaust and Lung Cancer Risk in a Pooled Analysis from Case-Control Studies in Europe and Canada".

Bisphenol A Roundup

Since it's detected at low levels in 95% of us and since Americans have been exposed to it for more than 50 years you'd think someone would have noticed if exposure to bisphenol A (BPA) were responsible for widespread illness, deformity and death. Apparently not, at least not if the findings from a recent wave of BPA studies are to be believed.

The new findings are, in no particular order, that BPA: (a) damages sperm (b) inhibits the normal development of ovaries (c) alters brain development (d) causes premature birth (e) may be a carcinogen like DES (f) damages blood cells (g) activates breast cancer cells (h) impairs the body's defenses against colon cancer especially in women; and, (i) makes offspring anti-social and neurotic. And those are just a few of the "findings" published in the last two months. Obviously the world that existed before 1950 or so, before  BPA was everywhere used to seal bacteria out of food and dental cavities, had to have been a much healthier and more peaceful one. Alas.

Why Do Acquarians and Geminis Differ About the Efficacy of Bone Marrow Transplants?

Let's say that you had chronic myeloid leukemia (CML) and that your doctor recommended a bone marrow stem cell transplant. Let's also say that you decided to check out the literature so as to be better informed. When you did you found that in a well designed study of 626 CML patients a large and highly statistically significant difference in risk of death from the procedure was revealed. Indeed, those transplant patients with the astrological sign of Aquarius were likely not to die in the following 5 years whereas those born under the sign of Gemini had a much greater risk of death and in fact were more likely than not to die within 5 years of the procedure. Finally, let's say that you are, in fact, a Gemini. Do you go ahead with the transplant?

Ok, ok. Before you make up your mind you go looking for other papers published in respected peer reviewed journals that look at zodiac sign and medical treatments. Guess what? You find other papers also showing that your sign is associated with a statistically significant increase in risk of injury or death! Now what?

Well, the first thing you need to remember is if you're dredging data for statistically significant associations and you don't find any, well, all that proves is that you don't know what you're doing. Epidemiology works when you formulate a hypothesis and then test it. When you go dredging for data without having in hand a hypothesis to be tested, knowing that the laws of probability will invariably generate statistically significant yet spurious associations, you're just manufacturing "science"; discovering nothing and flaunting the scientific method in the process.

For a fun read about why Geminis ought not worry about their sign and how statistically significant associations, based on the most rigorous statistical analyses, can be be generated at will to support, if that's the word for it, the most absurd causal inferences read: "Sign of the Zodiac as a Predictor of Survival for Recipients of an Allogeneic Stem Cell Transplant for Chronic Myeloid Leukaemia (CML): An Artificial Association".

Is Benzene a Teratogen?

Back when I was an associate we had a refinery client with a benzene unit and the unit generated litigation along with the benzene. Interestingly, the union knew that benzene could cause fatal blood diseases back when the unit was built in the late 1950s. The union even sent down a benzene safety poster which was framed and hung in the control room. The operators were issued cartridge respirators for aromatics and had blood samples taken for testing on a regular basis. Far from being a worry, even after the Emergency Temporary Standard for benzene came out the unit was a favorite of those union members with the seniority to bid on to it. So much so that women with the requisite seniority would have hysterectomies so they could get around the so-called fertile female policy that prevented them from working on it.

Some early, and I might add poorly done, studies had suggested that benzene might cause birth defects. Accordingly many companies adopted policies that kept women who were otherwise qualified from working with materials or in operations feared to be teratogenic. Eventually the U.S. Supreme Court struck down such policies. See: International Union, United Automobile, Aerospace & Agricultural Implement Workers of America, UAW, et al v. Johnson Controls, Inc.

Nothing ever came of the claim that benzene was a teratogen and even in southeast Texas where injury and illness would otherwise not exist but for deep pocketed corporations we never had any benzene birth defect claims.

Now however there's a new study in Environmental Health Perspectives that compares EPA estimates of benzene exposures in Texas and the rates of spina bifida among people living in those areas during a recent five year period. It found for the area with the highest exposure a 2.3-fold increase in spina bifida; an association which was statistically significant.
Read all about it (free) in "Maternal Exposure to Ambient Levels of Benzene and Neural Tube Defects Among Offspring, Texas, 1999 - 2004"

Hat tip: The Houston Chronicle

Biases, Fear and Thyroid Cancer

In the NYTimes' "Consults" today an endocrinologist answers questions about whether thyroid cancer has to do with toxins or genes. The first question from "Toxins, Genes and Thyroid Cancer" inquires as to whether a cancer "cluster" in an office could be due to environmental (i.e. man-made) toxins. The endocrinologist correctly responds that the incidence of thyroid cancer has increased dramatically over the last two decades. He goes on though to hedge that "no specific chemical or environmental factor has been demonstrated to commonly cause thyroid cancer in humans." With regard to whether the illness is hereditary he takes a firmer position stating that most forms of the cancer are not considered to be so. Only three wrote in to comment. They wondered whether too much iodine, "hundreds of chemicals in our bloodstream" or flame retardants and the linings of cat food tins might be to blame.

If you defend toxic tort litigation or have ever tried to explain the sudden rise and then plateau in rates of breast or prostate cancer you know it's time to pause and beat your head on the desk for a bit.

Feel better? Now for something to fit with the other puzzle pieces. Read: "Increasing World Incidence of Thyroid Cancer: Increased Detection or Higher Radiation Exposure?" As the authors make clear, improved detection and screening may well be, but has not quite yet been confirmed to be, the answer. In any event the evidence for better detection is a whole lot stronger than for cat food tins and copy machines.

Listeria Monocytogenes Outbreak Traced to Celery; Or Was It?

 

The Texas Department of State Health Services (DSHS) has shut down a San Antonio packager of produce for restaurants and schools following its determination that a L. monocytogenes outbreak thought to have been responsible for five deaths of immune compromised individuals originated at its facility. The company targeted for the shutdown is disputing the claim saying that its testing showed no L. monocytogenes and questioning whether the samples taken from which the bacterium was cultured were contaminated thanks to shoddy procedures by DSHS personnel. Read about it in "San Antonio Produce Plant Closed by Health Agency" in the Houston Chronicle. The DSHS news release can be found at "DSHS Orders SanGar Produce to Close, Recall Products".

L. monocytogenes contamination of ready-to-eat foods is a world-wide problem. See: "Occurrence of Listeria Monocytogenes in Ready-To-Eat Foods From Supermarkets in Southern Italy". For a recent discussion of molecular markers for detection and attribution and biomarkers for surveillance and epidemiological investigations see: "Future Challenges to Microbial Food Safety".

Beryllium Exposure Associated With Several Types of Cancer

A follow-up of workers at beryllium processing plants showed significant associations with lung cancer, cancers of the nervous system, urinary tract cancers and chronic obstructive pulmonary disease. Most diseases of interest showed a dose-response relationship and peak dose relationship. See: "Cohort Mortality Study of Workers at Seven Beryllium Processing Plants: Update and Associations With Cumulative and Maximum Exposure".

Getting the Causation Cart Before the Benzene Horse

Let's assume you're trying to prove that benzene causes a host of cancers of the hematopoietic system - essentially all lymphohematopoietic neoplasms. Wouldn't it be clever to argue that the best studies are those that confirm your bias; i.e. that benzene just has to cause acute nonlymphocytic leukemia (ANLL, AML, etc.)? Better yet, wouldn't it be really clever to enforce plaintiffs counsels' demand that, thanks to a little ex post reassessment, studies including subjects prospectively assessed to have been exposed but after the fact, in an instance or two, found to have spent more time in an office, be marginalized or excluded outright? Finally, wouldn't it be brilliant to assume that those with an increased risk of AML just had to have had higher exposures thereby bringing all their similarly situated brethren into the study along with them?

Well, when you don't do experiments but merely look at the literature and run statistical tools over the top of those papers you cherry pick until the stars align (i.e. show that benzene causes everything hematopoietic) it ought not be a surprise that you wind up verifying your preconceptions. And such was the case with "Occupational Benzene Exposure and the Risk of Lymphoma Subtypes: a Meta-Analysis of Cohort Studies Incorporating Three Study Quality Dimensions".

If benzene was indeed some sort of universal blood carcinogen you'd think the statistical evidence wouldn't need to be so tortured before yielding.

Schools: Highly Effective Virus Transmission Sites

Two weeks after kids go back to school, doctor visits for flu-like symptoms spiked in 2009. Why? It looks like schools catalyzed "community-wide transmission" of H1H1 pandemic influenza. See "School Opening Dates Predict Pandemic Influenza A(H1N1) Outbreaks in the United States". The authors suggest vaccination begin before kids return to school.

There's good news for people (like me) who hate getting vaccinated. It looks like "Needle-Free Influenza Vaccination" is on the way.

Finally, here's the CDC's picture of the week from the Public Health Information Library. It's what influenza vaccines help your immune system recognize and inhibit.

PHIL Image 10073

Citing a 16% Increase in Risk of Major Adverse Cardiovascular Events FDA Recommends Meridia Use be Discontinued

New data from the Sibutramine (Meridia) Cardiovascular Outcomes (SCOUT) trial showed little weight loss but significant increased risk of non-fatal heart attack, stroke, a cardiovascular event requiring resuscitation and cardiovascular death. The FDA suggests that those taking Meridia for weight loss stop taking it, speak to their physician about alternate treatments and contact a physician "right away" should they experience chest pain, palpitations, etc. The FDA recommends physicians stop prescribing Meridia, contact patients currently on Meridia and ask that they stop taking it, inform patients of these risks, assess patients for evidence of outcomes associated with Meridia use and to report any side effects to the FDA's MedWatch program.

Abbott Laboratories has agreed to withdraw Meridia and advises patients to discontinue the use of sibutramine and to consult their physician for alternatives. Abbott's press release contains an extensive discussion of the risks, what patients need to do and the potential side effects of which they ought to be aware. For additional information see www.sibutramine.com .

Given what's happened of late with any drug remotely associated with the treatment of Type II diabetes I'll go out on a limb and predict that companies will in the future be very wary of developing treatments for this malady or any underlying factor such as obesity.

Chronic High Noise Exposure and Coronary Heart Disease: A Strong and Consistent Association

In "Exposure to Occupational Noise and Cardiovascular Disease in the United States: the National Health and Nutrition Examination Survey 1999 - 2004"  researchers report a doubling or tripling of the risk for coronary heart disease among workers chronically exposed to high occupational noise. Meanwhile, other studies are showing that men subjected to aircraft noise from nearby airports and overflights face an increased risk of hypertension. See "Aircraft Noise and Incidence of Hypertension-Gender Specific Effects", "Exposure-Response Relationship of the Association Between Aircraft Noise and the Risk of Hypertension" and, for evidence that helicopter noise is most strongly associated with an increased risk of hypertension see "The Effects of Chronic Exposure to Aircraft Noise on the Prevalence of Hypertension".

Here are a few Power Point presentations (.pdf format) from the WHO on the topic of the health impacts of chronic noise exposure:

"Noise and Children"

"Quantifying Burden of Disease from Environmental Noise"

"Health Effects of Noise"

All Things in Moderation: Drinking While Pregnant Edition

Does the dose make the poison? In toxic tort litigation plaintiffs have long argued that at the unmeasured and unobserved low dose end of the dose-response curve risk doesn't reach zero until the dose reaches zero. To support their claim they point to regulators' linear no-threshold risk models, they try to throw the burden of proof on defendants and they conclude by saying that since defendants can't prove there's an absolutely safe level for exposure, all levels must therefore be unsafe.

This "no safe level" argument isn't confined to cancer cases and plaintiff lawyers aren't the only ones who make it. Advocacy groups for a variety of human ailments stake out similarly extreme positions. For example, March of Dimes claims that some 40,000 American children are born annually with fetal alcohol syndrome disorders (FASDs). In addition to claiming "no level of alcohol use during pregnancy has been proven safe" they cherry pick data from weak studies to assert that mothers who consume as little as one alcoholic drink per week have children with (a) small heads ("a possible indicator of brain size"); (b) a 300% increase in risk of growing up to be delinquents; (c) a variety of emotional and learning disorders; and, (d) an increased risk of becoming alcoholics and drug addicts. Finally, March of Dimes flatly states "[t]here is no cure for FASDs."

The good news is that there never has been much evidence to support these horror stories and the better news is that there's a brand new study showing that not only are the children of light drinkers at no increased risk of cognitive defects (at least through age 5), they're likely to have fewer problems, be less prone to hyperactivity disorders and have higher cognitive test scores. See "Light Drinking During Pregnancy: Still No Increased Risk for Socioemotional Difficulties or Cognitive Deficits at 5 Years of Age?"

There's no doubt that chronic binge drinking during pregnancy can do lasting harm to a woman's fetus. Similarly, there's no doubt that roasting yourself in the sun all summer and continuing to irradiate yourself in a tanning bed the rest of the year can lead to malignant melanoma. Yet extrapolating from such findings to declare that there's no safe level of exposure to sun or alcohol or whatever not merely panics parents needlessly; it may well result in the infliction of needless harm on the very people for whose benefit such advocacy is intended.

Avandia: A Fair Cop?

Without a randomized controlled trial we may never really know the answer to that question though maybe the complex systems / molecular biology people will get it worked out one of these days.

We will however get to find out if the data on which the RECORD trial (Rosiglitazone Evaluated for Cardiovascular Outcomes in Oral Agent Combination Therapy for Type 2 Diabetes - don't ask me how that works out to "RECORD") was based had been properly characterized before being analyzed. In "Regulatory Action on Rosiglitazone by the U.S. Food and Drug Administration", the announcement of new tighter restrictions on Avandia and the end of the TIDE trial, it was also disclosed that the FDA "is requiring an independent readjudication of end points at the patient level" of the RECORD trial's data. While the RECORD trial apparently established the noninferiority (NI) of Avandia versus the standard treatment, claims of design flaws that biased the results in favor of Avandia, data cherry-picking, and worse were subsequently raised (see "The Doctor Doth Protest Too Much, Methinks") .

What all this dredging and now reclassification-to-be-followed-by-redredging will reveal is anyone's guess and at best it will only lead to a posthumous vindication. The more important question is how will the battle between those who hold epidemiology to be a source of actionable knowledge and those who hold it to be a source of testable hypotheses play out. On one side of the debate, at least in the Avandia context, are idealistic academics. Their presentations to FDA were filled with lots of exclamation points, suspicions and worries about the rich and powerful exploiting the poor, sick and vulnerable. On the other side are the practitioners, pragmatists and industry scientists. They (to ruin my metaphor) played Spock to to the anti-Avandia group's Captain Kirk.

Will Avandia wind up with a pardon? If so, this episode may in hindsight turn out to have been the last stand for a methodology that for 40 years reliably generated the "knowledge" needed to support a particular world view; a methodology which has now gone decades without a big hit (e.g. asbestos). If not, and should Dr. Nissen's view prove to be the accepted one, the conclusion that the approval of Avandia was "one of the worst drug safety tragedies in our lifetime" resulting in up to 47,000 needless heart attacks could reignite a "science" that for three decades ushered in wave after wave of mass tort litigation. Whichever way it turns out I'm guessing it'll be Avandia's lasting legacy.

 

Vitamin D and Cancer Prevention: Hold the Applause

To test the hypothesis that vitamin D plays a role in protecting people from cancer wouldn't it be sensible to monitor the serum levels of the vitamin in thousands of subjects and see if it's related to their risk of cancer? It's been done and the results are in.

"Serum Vitamin D and Cancer Mortality in the NHANES III Study (1988-2006)", the work of the NCI's Division of Cancer Epidemiology and Genetics, reports that whatever the levels, whatever the latitude and whatever the race among 16,819 participants across the Americas there was no evidence that vitamin D prevented cancer.

Of course, epidemiology being epidemiology, if you want good news on vitamin D and cancer all you have to do is open the next journal and find "Vitamin D in Health and Disease" and you can feel better about taking your vitamins as it reports, upon reviewing the literature, a protective effect not only from breast and GI cancers but heart disease, diabetes and more.

And so it goes.

 

Trichloroethylene + Gene Variant = Renal Cancer?

Workers exposed to trichloroethylene (TCE) who carry at least one copy of the GSTT1 allele are reported to have an 88% increase in risk of renal cancer in the new paper "Occupational Trichloroethylene Exposure and Renal Carcinoma Risk: Evidence of Genetic Susceptibility by Reductive Metabolism Gene Variants." Those workers without the polymorphism had a slight decrease in risk. Given that the allele occurs on a gene coding for cysteine β-lyase, which plays an important role in the metabolism of TCE among other molecules, the finding demonstrates biologic plausibility as well as increased risk.

So back to yesterday's post about risk : which risk, if any, would be relevant in a TCE toxic tort case? The risk given to all workers collectively; the risk at a particular range of exposure; the risk given to those carrying the polymorphism; or, the risk to those with the polymorphism exposed at high levels? And could it be the case that one risk is relevant to the question of whether a defendant's conduct was reasonable while another was relevant to the question of causation? How would that work?

However it works, as the causes of individual susceptibility are identified expect these sorts of challenges to multiply. 

TIDE Trial on Hold; Likely Never to be Completed

That's it for Avandia. The FDA has put the TIDE trial "on full clinical hold and the regulatory deadlines for its conduct are rescinded." The plan is to reassess the data from the RECORD trial to see if reliable "ischemic risk can be obtained from [] re-adjudication..." In Europe physicians have been directed to stop prescribing the drug. Meanwhile, in the U.S., patients will still be able to get the drug if they are already on it or have tried other drugs like pioglitazone without success.

Coffee, Tea or Brain Cancer?

Glioblastoma multiforme is an aggressive and invariably fatal primary brain cancer. The cause(s) of glioblastoma is (are) unknown but one factor is strongly and consistently associated with a decreased risk of the cancer. That factor is the consumption of caffeinated tea and coffee; and the association is dose-dependent. Men were reported to have a 41% decrease in risk while women experienced a 26% decrease in "Coffee and Tea Intake and Risk of Brain Tumors in the European Prospective Investigation into Cancer and Nutrition Cohort Study" just published in The American Journal of Clinical Nutrition. Those numbers are consistent with the findings reported in "Coffee, Tea, Caffeine Intake, and Risk of Adult Glioma in Three Prospective Cohort Studies".

What's with all the good news about coffee and tea? It stands in stark contrast to the epidemiology published in the 1970s and early 1980s. Back then the literature was full of reports linking coffee in particular to pancreatic cancer, kidney cancer, colon cancer, diabetes, heart attack and bladder cancer. The difference perhaps is due to the fact that modern studies involve huge cohorts and assessment of exposure (dose) is much more rigorous. Yet what should be made of all the big studies of non-coffee drinking Seventh-Day Adventists and their decreased risk of all manner of maladies? Only that something besides skipping coffee is at work.

When You Have to Prove a Doubling of the Risk Ask "For Whom and During What Time Period?"

Despite the fact that the Texas Supreme Court in Merrell Dow Pharmaceuticals, Inc. v. Havner wrote "[w]e do not hold, however, that a relative risk of more than 2.0 is a litmus test..." many lawyers and some courts believe that Havner (and even Daubert) require that plaintiff establish epidemiological evidence of a "doubling of the risk" before she can establish so-called specific causation. In my view all those two courts ever said was that if you wanted to play the game of epidemiological causal inference then you had to play by epidemiology's rules; and, furthermore, that if all you had was probabilistic evidence then that evidence had better show that defendant's product probably did it.

Whatever the interpretation, we mass tort lawyers often wind up fighting over whether there has been a doubling of the risk. One thing we're just beginning to fight over is "when do you measure the risk?" Take for instance the Women's Health Initiative and hormone replacement therapy (HRT).

Depending on when you decide to look HRT either caused an 80% increased risk (after one year of treatment) of coronary disease or a 30% decreased risk (after five years) of coronary disease. So how do you choose which risk is the "real" one?

In what promises to become an epidemiology blog posted for free at Epidemiology is a copy of "The Hazards of Hazard Ratios". In it the author makes the point that hazard ratios, an approximation of risk ratios, often vary over time and may be subject to biases, as when, hypothetically, over time the exquisitely sensitive, by virtue of suffering the malady earlier, manage to be deselected from subsequent years' calculations.

The phenomenon is well known to those doing benzene litigation. Had the study of the Pliofilm workers been done today the very same cohort would demonstrate a relative risk less than 2.0 quickly trending towards 1.0. The obvious retort is that "latency" somehow is responsible but that doesn't explain the fact that no member of the cohort has developed leukemia in decades. The most sensible answer is that those susceptible of getting it get it and those who aren't don't just fail to get it - they can't.

In other words, the distinction in toxic torts between "general causation" and "specific causation" is likely often (if not usually) a false one. Thanks to the laws of physics, bullets can reliably be said to be a general cause of bullet holes in people since all are susceptible to the effect. The same cannot, however, be said of benzene or HRT. Apparently they only increase the risk (perhaps to 1.0) in those people who, due to genetics, epigenetics, microbiota, environment or some other factor(s), are primed to produce the effect while simultaneously imposing an increase of 0.0 on everyone else.

So how, other than flawed post hoc "reasoning", do we determine whose injury was caused but for an exposure? Biomarkers have clearly failed of their promise. Now what? The way things have been going I bet it'll look something like this: "Discovering Graphical Granger Causality Using the Truncating Lasso Penalty" but maybe we'll get lucky and it'll look more like: "Causal Diagrams and Change Variable". In the meantime plaintiffs without biomarkers need to do a better job of demonstrating how they are like those in the time interval with the biggest risk.

Cancer Rates Among Cosmetologists

Remember "The Story of Cosmetics" - Sophistry At Its Finest? Well, there's a new epidemiological study of cosmetologists that throws cold water in the face of the cosmetics scare. In "Cancer Incidence in Female Cosmetologists and Manicurists in California, 1988 - 2005" researchers report a 16% decrease in the cancer incidence among 325,228 licensees. Even when sorted by cancer site no large increase for any cancer could be found though small increases were noted for thyroid and lung cancer.

Avandia's Revenge

The Avandia witch hunt saga just gets more and more absurd. Guess what just popped up from the FDA's Division of Drug Information? Notice of an ongoing safety review of pioglitazone (the "safer alternative" generic pushed by those urging Avandia be pulled from the market.). Apparently "an increased risk of bladder cancer was observed among patients with the longest exposure to Actos (pioglitazone) as well as in those exposed to the highest cumulative dose of Actos."

So does this mean that pioglitazone ought to be banned? Does this mean that the ongoing study of pioglitazone users needs to be discontinued because it's unethical? That's what the anti-Avandia activists argued re: Avandia and the TIDE trial; they wanted everyone off Avandia, on pioglitazone and the experiment halted - pronto.

No; it ought not mean any of those things. What it does mean is that it's foolish to let our decisions be bound by the results of a single data dredge.

For more on the Avandia issue see: "Avandia: Burn Her Anyway?" For more on the perils of data mining see "Lies, Damned Lies and P-Values". Here's the full text of today's FDA missive on pioglitazone:

 

FDA/CDER/Division of Drug Information (DDI)

The Division of Drug Information (DDI) is CDER's focal point for public inquiries. We serve the public by providing information on human drug products and drug product regulation by FDA.


The U.S. Food and Drug Administration (FDA) is reviewing data from an ongoing, ten-year epidemiological study designed to evaluate whether Actos (pioglitazone), is associated with an increased risk of bladder cancer. Findings from studies in animals and humans suggest this is a potential safety risk that needs further study.

Actos is used along with diet and exercise to control blood sugar or improve control of blood sugar in adults with type 2 diabetes mellitus.

Bladder cancer is estimated to occur in 20 per 100,000 persons per year in the United States and is thought to be higher in diabetics.

The drug manufacturer, Takeda, has conducted a planned analysis of the study data at the five-year mark, and submitted their results to FDA. Overall, there was no statistically significant association between Actos exposure and bladder cancer risk. However, further analyses were also performed looking at how long patients were on Actos and the total amount of the drug they received during that time. An increased risk of bladder cancer was observed among patients with the longest exposure to Actos, as well as in those exposed to the highest cumulative dose of Actos.

At this time, FDA has not concluded that Actos increases the risk of bladder cancer. Its review is ongoing, and the Agency will update the public when it has additional information.

For more information, please visit: Actos

 

Parkinson's: Not Much Evidence for Manganese; Strong Evidence for Vitamin D Deficiency

In Tamraz v. Lincoln Electric Company, et al the 6th Circuit held that an expert could not stack speculation about mechanisms upon unseen and undetected lesions in the plaintiff to get from manganese exposure to his Parkinson's disease. Furthermore, the court ruled that you can't reasonably arrive at a causation opinion using what's erroneously called a differential diagnosis (it's more properly called simply a process of elimination) when you've no sound basis for ruling anything either in or out.

On the other hand, there's growing evidence that vitamin D deficiency is strongly associated with Parkinson's. Best of all, it makes sense. It looks like the enteric nervous system is first to be degraded in Parkinson's and that system is exquisitely sensitive to the gut mediated immune system which in turn is adversely affected by vitamin D deficiency. See: "Parkinson Disease: Could Sunlight Offer Protection from Parkinson Disease?"

A New Formaldehyde and Leukemia Meta-Analysis

Martyn Smith et al have just had published "Formaldehyde and Leukemia: An Updated Meta-Analysis and Evaluation of Bias" in the Journal of Occupational and Environmental Medicine. They report a relative risk of 2.47 for myeloid leukemia for those with high occupational exposures to formaldehyde.

Meat and Colorectal Cancer: Good News

All things in moderation. That wisdom applies profitably to meat consumption; including the processed variety. See "Meat, Poultry and Fish and Risk of Colorectal Cancer: Pooled Analysis of Data From the UK Dietary Cohort Consortium".

A Two-Fold Increase in Lung Cancer Risk Among Mexican-Americans Exposed to Pesticides

Researchers at MD Anderson report that pesticide exposure in general was associated with a doubling of the risk of lung cancer among 212 Mexican-American lung cancer patients at the hospital. The exposure data was, however, troublingly anecdotal and unverified. See "Assessing Environmental and Occupational Risk Factors for Lung Cancer in Mexican-Americans".

Another Large Study Finds No Increased Risk From Avandia

In "Risk of Cardiovascular Events and All-Cause Mortality in Patients Treated With Thiazolidinediones in a Managed-Care Population" the authors report on the health outcomes of over 36,000 patients with type II diabetes taking either Avandia or pioglitazone. Over five years 4.16% of the patients on Avandia suffered heart attack, heart failure or death while 4.14% of those on pioglitazone suffered similar outcomes. The difference was obviously insignificant.

Well aware of the controversy into which the journal has waded and the shrill ad hominem attacks launched against anyone suggesting that the drug doesn't increase the risk of heart attacks, heart failure or death a free editorial accompanies the study. It confesses that the evidence against Avandia is equivocal at best, frets about the find it and bind it approach to supposed biomarkers and then limply concludes "[t]hus, the simplest message might suggest the avoidance of [Avandia] until more reassuring evidence for the drug becomes available." What that evidence might be, or how much of it there ought to be, isn't discussed. I'm guessing that the hope here is that the TIDE trial is halted so that embarrassing questions don't ever have to be answered.

How Quickly Do Bacteria Evolve? Lots Faster Than You Thought.

It has generally been assumed that bacteria evolve to overcome host defenses at a rate much slower than that of viruses. That assumption appears to be wrong. Helicobacter pylori, a cause of multiple cancers in humans, mutates at a rate similar to many viruses. See: "Microevolution of Helicobacter pylori During Prolonged Infection of Single Hosts and Within Families".

Just how fast can bacteria evolve? Would you be surprised to learn that within the lifetime of mice, bred to be free of gastrointestinal bacteria, bacteria from drinking water cannot only set up shop and colonize their hosts' guts but evolve into whole new species? If you are surprised, read: "Bacteria From Drinking Water Supply and Their Fate in Gastrointestinal Tracts of Germ-Free Mice: A Phylogenetic Comparison Study".

The long war continues, of course; and while we've been basking in our assumed victory over them for the last forty years our ancient enemies, descended from a long line of brilliant sappers, have been hard at work.

Painters Have an Increased Risk of Bladder Cancer

According to a new meta-analysis of studies of painters there's a small but consistent increase in the risk of bladder cancer among painters. What it is about painting that accounts for the risk is unknown. See: "Bladder Cancer Risk in Painters: a Meta-Analysis".

Eco-Epidemiology and the Early Puberty Panic

Before the paper was even pre-published online the lead author of "Pubertal Assessment Method and Baseline Characteristics in a Mixed Longitudinal Study of Girls",  Dr. Frank M. Biro, was giving interviews. The measured write up in The New York Times reported that "[s]ome researchers also suspect that environmental chemicals that mimic the effects of estrogen may be speeding up the clock on puberty, but that idea is unproved." Later however it quotes Dr. Brio, whose new project involves an attempt at correlating the subjects' hormone levels and chemical exposures, as saying "I think we need to think about the stuff we're exposing our bodies to and the bodies of our kids. This is a wake-up call, and I think we need to pay attention to it." From what I can see of the article it doesn't even purport to examine such an association and nothing about the data other than a crude sort of - our use of more and different chemicals may be going up (an assumption which is not assessed in the article) and onset of one prepubertal sign is going down - would even suggest such a hypothesis.

MSNBC took a far less measured approach. Whereas the NYTimes correctly reported that the study actually found no change in the onset of menses, MSNBC reported that "American girls are maturing earlier and earlier." Then it went on to write about all the possible perils of those who "go through puberty early" without disclosing that the 7 and 8 year old subjects had not in fact "gone through puberty" at all. And while MSNBC discusses the putative link between obesity and early puberty they quickly move on to a comment from Dr. Stanley Korenman (not one of the authors) who "says environmental exposure to estrogens in plastics, chemicals and foods has been going up" adding "and estrogens do stimulate breast development". All of this then prompts Dr. Biro to conclude "[u]ntil we know what the cause is, the best way to slow puberty may be to 'start living green'.

What's behind this belief that there's a simple answer to something like trends in the onset of puberty? Well, there's that whole third age of epidemiology business whereby many came to believe that changed and chronic conditions just had to be man-made. And there's also risk factor epidemiology which, if done often enough, is essentially guaranteed by the laws of probability to produce a study showing a statistically significant association between condition and chemical. Of course, as we've written before, that now 40 year old paradigm and the tool used to support it have been crumbling under the weight of erroneous predictions for the last 20 years. The problem was that causation was far more complex than most people expected and many of the tools needed to model causal networks had yet to be developed.

Enter eco-epidemiology. For a discussion of this new field you'll not find a better primer than "The Eco- in Eco-Epidemiology" and if you're really really interested and want to watch some lectures on the topic try The Yale Center for Eco-Epidemiology. For those beginning to doubt that nature is benign try the talks from "Zoonoses: Diseases from Nature". Finally, for those wanting the short version suffice it to say that causation, even for something like AIDS, is very very hard. A host of impacts from genes to culture determine who and how many will be infected. And genes shape cultures and culture may well shape genes . (h/t GNXP) Eco-epidemiology aims to deal with all of these multilayered causes.

Finally, back to the puberty panic. Our culture has changed in many ways over the last fifty years. For example resources have become more abundant. Elsewhere in nature when that happens critters exploit the opportunity by breeding sooner and more often. Why should we assume that we don't, albeit unconsciously, do the same? Eco-epidemiology aims to find all the contributors to the human condition and let the chips fall where they may.

Better Have Another Cup of Joe

Coffee, especially caffeinated coffee, has been the subject of numerous epidemiological investigations; and the endless cycles of "this week scientists said it causes cancer, last week they said it doesn't and next week they'll probably say it prevents it" did as much as anything to make even epidemiologists ask whether it was time for the science of black box epidemiology "to call it a day". Of course they didn't call it a day and many continue to try to discover whether everyday "stuff" like coffee and tea are in fact deadly carcinogens. They do it by attempting to discern an association between exposure and disease beyond what chance would predict. Here's the newest one on coffee: "Coffee and Tea Intake and Risk of Head and Neck Cancer: Pooled Analysis in the International Head and Neck Cancer Epidemiology Consortium".

The study is large and looks to be very well done. Overall, neither coffee nor tea seems to be associated with head and neck cancers but what's interesting is the unexpected and dose dependent decrease in cancer associated with drinking lots of caffeinated coffee each day. Those downing three or four cups each day reduced their risk by 30% and those drinking more than four cups per day lowered their risk by almost 40%. What to make of it?

Well, as we regularly argue here, it's pretty much impossible to know what to make of it. We could jump to the conclusion that coffee battles cancer but we'd be ignoring, among a host of other factors, the fact that a lot of head and neck cancers are caused by human papillomavirus (HPV). The authors write that "there is no reason to think that coffee intake is associated with human papillomavirus infection" but just because they can't think of a reason doesn't mean there isn't one. Maybe heavy coffee drinkers have fewer partners (all work and no play ... etc.). Maybe some people have a rare gene that simultaneously protects against head and neck cancer and makes them crave caffeinated coffee. Who knows?

None of which is to say that these sorts of studies are useless. In fact they're often invaluable. The point is that such studies ought to be put in perspective. They aren't valuable because they answer the question of causation. They can't. They're valuable because they generate insightful clues that can be shaped into testable hypotheses that if borne out lead to real answers about causation and from there to further clues about prevention and treatment.

PM2.5: Diabetes, Heart Attack, Lung Cancer, Premature Death, etc

Etc indeed. The list of maladies laid at the feet of inhaled particulate matter smaller than 2.5 micrometers (thus PM2.5) is long and growing. You can add diabetes to the list thanks to "Association Between Fine Particulate Matter and Diabetes Prevalence in the United States" and lung function deficits in early childhood too ("Effect of Prenatal Exposure to Fine Particulate Matter on Ventilatory Lung Function of Preschool Children of Non-Smoking Mothers").

Is it a certain type of ultrafine particle that's responsible? Some studies say yes and others say no. In vitro toxicity testing tends to suggest that altered function is due simply to particle size while epidemiology studies tend to cast blame on one sort of particle rather than another though the findings vary from study to study and often conflict (a common problem when looking for weak effect associations). Do the observed effects meet the so-called specificity criterion for causal inference? At first the reported ill effects of exposure were said to be cardiovascular but now everything's in play especially since several studies have linked PM2.5 and Premature Death - All Causes.

So, is PM2.5 a universal toxicant and among the leading causes of death? Or could it be that people who live in urban areas with higher PM2.5 levels tend to have higher rates of unhealthy living? Is there anything good to be said about PM2.5? For example, why do farmers, who are often exposed to high levels of PM2.5, especially from endotoxins (think bits of bacteria), often have lifelong protection from many allergies that afflict those exposed to lower doses?

There's also the question of what's to be done about PM2.5. Farmers produce lots of it what with their gravel roads, grain bins, diesel tractors and plowed fields. EPA intends to regulate PM2.5 down on the farm and much more strictly than in the past but at what cost? And for those who don't like cost-benefit analyses what if the changes needed to reduce farm PM2.5 simply causes generation of ultrafine dust to be shifted elsewhere; and to increased markedly? See "The Environmental Cost of Reducing Agricultural Fine Particulate Matter Emissions".

Finally, of course, there's the issue of whom shall be sued. The finding that a speck of cotton dust from your shirt is as toxic (or as non-toxic, depending on how things shake out) as soot from combusted diesel fuel is an obvious impediment to to the diesel litigation plus there's a new study of truck drivers demonstrating that their presumed PM2.5 mortality may not be due to their work but rather can be, at least in part, explained by ultrafine dust exposures in and around the home: "Long-Term Ambient Multi-Pollutant Exposures and Mortality". Efforts to target other deep pockets will have to wait until science produces more definitive answers about what's to blame and how it can be determined that the PM2.5 in question was the cause in fact of the plaintiff's demise - likely an impossible task since causation in such circumstances is almost certainly the result of a constellation of factors; a constellation to be explored by something called eco-epidemiology. More on that another day.

What's Behind the Rise in Food Allergies?

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?

A Role For Observational Epidemiology?

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.

Twenty Suspected Carcinogens

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.

The Doctor Doth Protest Too Much, Methinks

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.

Avandia: Burn Her Anyway?

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.

NHL is Not Associated With Exposure to Gasoline

Twenty one of twenty two studies in the English literature that have examined the risk of non-Hodgkin lymphoma (NHL) among petroleum workers exposed regularly to gasoline have found no association between the two. The overall relative risk across all such studies has been estimated to be just 1.02. See: "Occupational Exposure to Gasoline and the Risk of non-Hodgkin Lymphoma: A Review and Meta-Analysis of the Literature".

What Percentage of Mesothelioma Cases Are Due to Asbestos?

In Japan it's 79.2%.

What Are The Risk Factors For Lung Cancer Among Those Who Never Smoked?

It doesn't appear to be second hand smoke; at least not from a brand new study of lung cancer cases in Toronto. On the other hand, having a family member diagnosed with lung cancer before age 50 significantly increased risk. So did a prior history of chronic lung disease. Similarly, occupational exposure to solvents, paints, thinners, soot, wood dust, grain dust and welding more than doubled the risk of lung cancer among non-smokers. Read all about it (for free) in "Lung Cancer Risk in Never-Smokers: A Population-Based Case-Control Study of Epidemiologic Risk Factors".

More Fuel for the Glargine Insulin Fire

A year ago a large European study implicated high doses of glargine insulin in certain cancers in diabetics. At the time endocrinologists urged causation in interpreting the results and advisory boards urged patients and physicians not to change their insulin regimens until further studies were completed. Now the results of those studies are starting to be published.

In "Doses of Insulin and its Analogues and Cancer Occurrence in Insulin-Treated Type 2 Diabetic Patients" the authors report the results of a nested case-control study of insulin-dependent diabetics. A five fold increase in cancer was found among those taking high doses of glargine. No such increase in risk was found among those diabetics taking other forms of insulin.

Vitamin D Deficiency and Multiple Sclerosis (MS): Who Pays?

In "The Lancet: Neurology" you'll find "Vitamin D and Multiple Sclerosis"  as well as "Vitamin D: Hope on the Horizon for MS Prevention?" Could it be that  the old mocked wisdom of "a healthy dose of sunshine" wasn't so silly after all? Could it be that the health panic precipitated by activists who demanded everyone stay out of the sun actually caused horrific and needless suffering? Could be.

I just got back from a vacation in Destin, Florida. While there I learned that there are people who cover every exposed surface of their kids with zinc oxide before letting them out in the sun. These parents think they're doing the right thing. Their kids seemed about as happy as I'd have been sent out into the world in a powder blue leisure suit. But while powder blue leisure suits don't cause rickets and MS, vitamin D deficiency does. When disease strikes will there be a viable cause of action against the scaremongers who caused it? It's an interesting question.

Why Would Mining Protect Men From Prostate Cancer?

Nobody knows why but miners tend to be at a much lower risk of prostate cancer. In "Could Mining Be Protective Against Prostate Cancer? A Study and Literature Review" researchers from Australia document a huge decrease in this most common of all cancers afflicting men. Before anyone hauls out the "healthy worker effect" note that the risk of prostate cancer in these miners, only 35% that of non-miners, is much lower than would be predicted even from those studies in which a significantly decreased risk has been attributed to the effect.

Poultry Workers at Increased Risk of Certain Cancers

Workers exposed to chicken viruses at slaughtering and processing facilities were at increased risk of dying from cancers of the pharynx, lung, pancreas and brain as well as from hematopoietic malignancies. See: "Cancer Mortality in Poultry Slaughtering/Processing Plant Workers Belonging to a Union Pension Fund".

Chronic Chlamydia Pneumoniae Infection and Lung Cancer

Smoking, asbestos, silica, hexavalent chromium, nickel and radon have all been blamed for cases of lung cancer. Now there's a new paper from the National Cancer Institute implicating C. pneumoniae: "Chlamydia pneumoniae Infection and Risk for Lung Cancer". Be sure to note the significance of the finding that it's the marker of chronic infection with which a significantly increased risk of lung cancer is associated.

Clostridium Difficile in the News

Can C. difficile be spread through the air? What does C. difficile have to do with COPD? How should physicians treat the new and especially virulent strain of C. difficile? At what temperature should you cook ground meat to kill C. difficile and its spores? Do alcohol-based gels kill it? Why do antacids administered in hospital increase the risk of infection? Are there new antibiotics that work against C. difficile?

Click on the links for a sense of current thinking on these issues.

An Unusual Benzene/MDS Opinion

In Quillen v. Safety-Kleen Systems, Inc., 2010 WL 2044508 (E.D.Ky.) the court determined that plaintiff's expert, Dr. George Rogers, could properly attribute a case of myelodysplastic syndrome (MDS) to benzene by doing a differential diagnosis. That some courts have taken to using differential diagnosis to identify the root cause of say splenomegaly rather than to distinguish histoplasmosis induced splenomegaly from Hodgkin's disease induced splenomegaly would likely set many physicians' eyes rolling.  Yet, that's apparently what the 6th Circuit said in Hardyman v. Norfolk & Western Railway Co., 243 F.3d255 (6th Cir. 2001) and thus the thinking by the Quillen court.

The point of doing a differential diagnosis, of course, is to rule out possible causes until just one is left - it's a process of elimination. But just because every other cause of splenomegaly has been ruled out in the case of a male patient that doesn't mean that it makes sense to conclude that the cause must be the remaining possibility - a metastatic ovarian cancer. To be considered for elimination in the first place the putative cause has to be one that makes sense. In Quillen though there was no effort to demonstrate that plaintiff's experience with benzene was the sort that would make benzene a reasonably plausible cause of his MDS.

Finally, please ponder the following. In response to the defendant's objection that plaintiff's expert had not ruled out ionizing radiation  the court wrote: "Defendant points to nothing in the record demonstrating that Quillen was ever exposed to a statistically significant amount of such radiation." Somewhere an epidemiologist just fell out of her chair.

Statins: A Little Sour With the Sweet

As evidence confirming the hypothesis that statins reduce the risk of cardiovascular disease piles up along with it comes data suggesting that some other risks may be elevated by taking statins. So far, however,  the benefits of statin treatment appear to easily outweigh the risks. Here's some of that data published in the last couple of weeks:

Lipid-lowering Agents and New Onset Diabetes Mellitus

Unintended Effects of Statins in Men and Women in England and Wales; Population Based Cohort Study Using the QResearch Database

Acute Renal Failure With the Combined Use of Rosuvastatin and Fenofibrate

Balancing the Intended and Unintended Effects of Statins

How is Multi-Drug Resistant Acinetobacter Baumannii Spread Through Hospitals?

It's not just unwashed hands nor even mostly unwashed hands. Gloves and gowns of healthcare workers are contaminated with A. baumanni more often than hands. Following contact with infected patients 38.7% of gloves/gowns were contaminated whereas hands only picked up the bacteria 4.5% of the time. A. baumannii looks to be more readily transmissible than even MRSA. See "Frequent Multidrug-Resitatant Acinetobacter Baumannii Contamination of Gloves, Gowns, and Hands of Healthcare Workers".

An Epidemic of Head and Neck Cancer

Cancer of the oropharynx in men is on the rise. Such cancers have traditionally been blamed on smoking and/or drinking so what accounts for its increase in a time of reduced rates of smoking and alcohol abuse? A virus; human papillomavirus (HPV). See "HPV-Associated Head and Neck Cancer: A Virus-Related Cancer Epidemic".

Remember, (1) the focus of this year's World Cancer Day was a call for greater awareness of the contribution of infectious disease to cancer; and, (2) the "age of receding panemics" never really passed - we just stopped looking for them.

 

Evidence For a Benzene-Leukemia Supralinear Dose Response?

What is the risk of leukemia in workers exposed to relatively low cumulative levels (<100 ppm yrs) of benzene? To get at the answer researchers examined nine published studies from which exposure-response data could be extracted. The exposure response curve that best fit the data shows a supralinear response, by which it is meant that the rate of response falls less rapidly for an interval before returning to its former rate of descent. The primary explanation offered for the deviation from the linear is that at some level the mechanism by which benzene is ordinarily metabolized is overwhelmed, or saturated, and metabolism by some other mechanism, one responsible for metabolites that cause leukemia, takes over until it too is saturated beyond which point risk rises at a lower rate.

The obvious problem with the best fitting curve graphed in the study is that it manages to predict a positive risk for benzene-induced leukemia even when the exposure to benzene is zero - an apparently absurd result. This best fitting model shows greater risk of benzene induced leukemias down to 10 ppm years than would be predicted by a typical linear no threshold model and a doubling of the risk level at just above 40ppm yrs. However, the second best fitting curve shows a sharp decline, especially below 40 ppm yrs and one which might be consistent with a threshold for risk at some non-zero level.

The paper is free and you can find it here: "Flexible Meta-Regression to Assess the Shape of the Benzene-Leukemia Exposure-Response Curve"

Libby Vermiculite and Death From Cardiovascular Disease

While examining the effect of cumulative fiber exposure on lung cancer and mesothelioma among Libby, MT vermiculite workers researchers from the ATSDR noticed dose related increase in deaths from cardiovascular disease. Using a "within-cohort comparison" they were able to demonstrate a significant dose response relationship at or above 44.0 f/cc years. See "Vermiculite Worker Mortality: Estimated Effects of Occupational Exposure to Libby Amphibole".

Breast Cancer and Diet: Good News and Bad News

in "Meat Mutagens and Breast Cancer in Postmenopausal Women - A Cohort Analysis" the researchers report the results of their study of breast cancer risk among women who consume meat cooked at temperatures high enough to produce heterocyclic amines (endocrine disruptors) and so-called "meat-derived mutagens". The good news is that the women were at no increased risk of estrogen-receptor positive breast cancer and at a slight decreased risk of estrogen-receptor negative breast cancer.

The bad news, if you've been forcing down lots of leafy greens in hopes of fending off breast cancer, is that cruciferous vegetable consumption seemed to have no effect, one way or the other, on breast cancer risk.

Was Estrogen-Progestin Hormone Replacement Therapy Responsible for High Rates of Breast Cancer in Marin County?

It certainly looks that way. Read: "Recent Trends in Hormone Therapy Utilization and Breast Cancer Incidence Rates in the High Incidence Population of Marin County, California".

There are three very interesting observations reported in the study. There's of course the strong correlation between the sharp drop in the incidence rate of invasive breast and the discontinuation of estrogen plus progestin hormone replacement therapy (EPHT). But there's also the fact that this group of well to do and well educated women apparently picked up on the HERS and WHI study results and almost overnight collectively and dramatically altered their preferences for a medical treatment. And then there's the discovery that hysterectomy rates varied wildly throughout the various counties in California.

As for the first finding, these and similar results continue to add to the overwhelming evidence that the breast cancer clusters in Marin County, CA and Long Island, NY are not due to environmental pollutants but instead are due to a mix of lifestyle, surgical procedures and medications along with a genetic component.

As for the second and third findings they reinforce the emerging view that comparing cancer rates from one county to those of other counties, to the state as a whole, or to national rates is perilous at best. Here Marin County women adopted EPHT early, did so at a high rate relative to women elsewhere in the state and underwent hysterectomies at a low rate relative to other women yet those causal factors were completely unaccounted for in early studies. As a result, activists and some researchers concluded that the cancer cluster had to do with the environment of Marin County rather than the lifestyle choices of its residents.

One of the underlying assumptions of epidemiological studies is that the population investigated and the group to which it's compared are the same. These results demonstrate that merely accounting for gender, race and age isn't enough and that lifestyle choices afforded by education and wealth can make a critical difference, both positive and negative.

But Sometimes Causation is Really, Really, Really Easy

As we wrote yesterday, causal attribution in most toxic tort cases is hard and the discovery that life and many of its diseases are emergent rather than predetermined phenomena has made the exercise even more complex. That said, sometimes causal attribution is easy - as in the cases of falling, being shot or developing mesothelioma after exposure to erionite.

50.5% of all deaths were due to mesothelioma in one village in Turkey and of women who moved from areas without erionite to this same village 69% of all deaths were due to mesothelioma. Read about it in: "Endemic Malignant Mesothelioma: Exposure to Erionite is More Important Than Genetic Factors".

Is there any carcinogen as potent as erionite? What do these numbers suggest about the mode of action of erionite? It sure looks as though the biological insult from erionite is more akin to that of some physical trauma than from some molecular biological disruption as in the case of a genetic mutation.

Huge Prospective Cohort Study Launched to Study Cell Phone Users

The Cohort Study on Mobile Communications (COSMOS) will follow more than a quarter milion people in Britain and Northern Europe over the next three decades to determine whether cell phone use poses any health risks. You can read about it at Reuters Health.

Given the accelerating changes technology has wrought over the last three decades I'd bet that a study of cell phones (at least the kind we currently use) will be as interesting to the people of 2040 as a study about the risks of using teleprinters would be to us today.

Does Shift Work Cause Cancer?

IARC thinks so. Others have recently published papers claiming that breast cancers and melanomas are probably caused by exposure to electric lighting at night. However, a new study published in the American Journal of Epidemiology looked at 73,049 Chinese women and found no association between breast cancer and night-shift work, irrespective of frequency, duration or cumulative shift work. The authors conclude that "it may be premature to consider shift work a cause of cancer."

Kidney Cancer: Cigarette Smoking, Obesity and Uncontrolled Blood Pressure

What causes kidney cancer? A lot of things have been proposed, especially substances and habits which are nowadays considered indicia of immorality such as using petrochemicals or having a Western lifestyle. However appealing to some, there's little to support such claims. On the other hand, smoking, being fat and not taking your blood pressure medicine are clearly the best ways to improve your odds of getting kidney cancer. Read about it in: "Contemporary Epidemiology of Renal Cell Carcinoma: Perspectives of Primary Prevention".

Another Study Concludes That Hormone Therapy Does Not Increase the Risk of Dying From Breast Cancer

See: "Hormone Therapy and Fatal Breast Cancer" finding no suggestive link between estrogen therapy and the risk of dying from breast cancer. Note that an increase in mortality from breast cancer among those taking both estrogen plus progestin could not be ruled out.

Lies, Damned Lies, and P-Values

In "Odds Are, It's Wrong", Tom Siegfried lays out the argument for the proposition that much of what you read in the scientific literature is wrong because many of the claims being made rely on statistical significance. You see, an impressive sounding statement like "the association between exposure and disease was highly significant (P<0.05)" does NOT mean (a) that there's a 95% chance that the association is causal; (b) that the absence of an association can almost certainly be ruled out; nor does it necessarily mean that (c) the finding is momentous, compelling or even important. It doesn't even say that if the test were to be repeated that its results would likely hold. A P-value, the arbitrary judge of "statistical significance", won't, and can't, have anything to say about the likelihood that a given hypothesis is or is not true.

The fact of the matter is that if you have a bunch of data and can't find at least one statistically significant association it in only proves one thing - that you're not trying hard enough. The magical P-value level of 0.05 is nothing but a trade-off; a balancing act between finding associations that don't exist (false positives) and missing true associations that do (false negatives). As a result, false associations are not only possible, they're guaranteed when you have enough data and slice it enough ways.

Now, lawyers are getting into the act. And while it's bad enough that "[a] lot of scientists don't understand statistics" (Steven Goodman quote from the "Odds Are, It's Wrong" article) it gets awful when lawyers try to deploy statistics to support or rebut claims. Law review articles are littered with claims resting on nothing more than small P-values. Some purport to show that certain appellate courts are biased against accident victims; others that tort reform is good for your health. And hardly a week goes by that I don't see a brief or a pleading asserting that Texas "jurisprudence" requires an epidemiological study with a risk ratio greater than 2 and a P<0.05 before a plaintiff can recover on a toxic tort claim. 

Apparently many lawyers, especially on the defense side, either forgot or never learned that it's easy to gin up false associations that meet the greater than 2 and less than 0.05 test. In fact, that's how most categories of toxic tort claims got started. Enshrining such a test in the law would turn out to be The Full Employment Act for toxic tort lawyers.

Causal inference from epidemiological statistical analysis is a crude method that nevertheless worked well for finding big effects like that of smoking on lung cancer risk and amphibole exposure on mesothelioma risk. On more subtle effects though, at the population level or molecular level, reliance on 20th century methods has produced so much bad science of late (bad only because statistics are routinely misused and abused and not because statistics aren't powerfully effective tools when properly used) that new methods of causal analysis are beginning to replace them. And these tools can answer the question of "how likely is it that drug A caused injury B?"

To see what the future of causal proof in toxic torts will look like read: "An Introduction to Causal Inference" by Judea Pearl.

Occupational Exposure to Endotoxins: A Good Thing?

In the newest edition of the journal Cancer Causes and Control you'll find a paper titled "Endotoxin Exposure and Lung Cancer Risk: A Systematic Review and Meta-Analysis of the Published Literature on Agriculture and Cotton Textile Workers". The authors examined 28 studies of workers occupationally exposed to high levels of endotoxins and their risk of developing lung cancer. Previous studies had suggested acute and chronic lung conditions could be caused by endotoxins.

Interestingly, endotoxin exposure was consistently associated with a large and statistically significant decrease in lung cancer. Furthermore, the protective effect was strengthened as dose was increased.

Also this month, in Cancer Epidemiology, Biomarkers & Prevention, you'll find "Lower Risk of Lung Cancer After Multiple Pneumonia Diagnoses". It turns out that getting pneumonia three or more times is even better than high exposure to endotoxins if you want to avoid lung cancer.

What is it about these biological challenges to the lung that leads to significant anti-lung cancer protective effect? It's anyone's guess but perhaps keeping your immune system tuned up is part of the answer.

No Association Between Industrial-Grade Talc and Mesothelioma

A new review of available toxicological, mineralogical and epidemiological data pertaining to talc mined in northern New York by R T Vanderbilt shows no support for the claim that exposure to it causes mesothelioma. See: "Industrial-Grade Talc Exposure and the Risk of Mesothelioma" just published in Critical Reviews in Toxicology.

Why Do So Many Non-Smokers Get Lung Cancer?

Read: "A Susceptibility Locus on Chromosome 6q Greatly Increases Lung Cancer Risk Among Light and Never Smokers". If you've got the gene your risk of lung cancer goes up nearly 500% even if you never smoke (in fact smoking only weakly increases your risk if you've got the gene). Interestingly, if you don't have the gene and you smoke your risk goes through the roof.

Everything is More Carcinogenic in China

"Tobacco smoking is responsible for one-third of the total cancer deaths among men." See "Estimation of Cancer Incidence and Mortality Attributable to Smoking in China". Secondary smoke exposure is said to be responsible for more than 11% of all cancer deaths in women.

Do you ever get the feeling that China is just going through what the U.S. went through in the 1960s, 1970s and 1980s when epidemiology and mass health panics were all the rage? If so, and if a "redistributional justice" model of tort law can be introduced in China, it may be deja vu all over again for classic mass tort litigation.

No Association Between Paint Fumes in the Home and Fetal Growth

See "Non-Occupational Exposure to Paint Fumes During Pregnancy and Fetal Growth in a General Population"

Though about half of the mothers surveyed said they'd been exposed to paint fumes in the home while they were pregnant the data suggested that the more fumes to which they'd remembered being exposed the lower the risk that their baby would be underweight. What? This study probably has more to say about the use of interview data as a proxy for exposure than it does about the relationship being examined.

Long Term Smoking Significantly Reduces the Risk of Parkinson's Disease

A greater than 40% decrease in Parkinson's if you smoke more than 30 years? So it seems from this huge NIEHS study of 305,468 Americans. "Smoking Duration, Intensity, and Risk of Parkinson Disease".

Obviously the risk of getting lung cancer, emphysema, etc from smoking is much higher. Still, if you knew you had the genes that protects you from lung disease (whatever they are) but not the ones that protect you from Parkinson's (whatever they are) would you smoke?

 

Are There Occupational Risk Factors for Parkinson's?

According to "Occupational Factors and Risk of Parkinson's Disease: A Population-Based Case-Control Study" the risk of Parkinson's was not significantly affected by workplace exposures to  metals, solvents or pesticides.

It Doesn't Seem Logical But It Does Seem To Be So

If people with Type 2 diabetes are at a greatly increased risk of heart disease wouldn't it make sense to get their blood pressure and triglycerides down and their "good" cholesterol up? Quitting smoking and lowering "bad" cholesterol reduces the risk from very, very high to just very high so attacking these other presumed risk factors should help, right? Besides, pushing systolic blood pressure down closer to normal would obviously yield some benefits. And there's no way it could hurt. Right?

It turns out that these interventions, implemented on the basis of reasoning and not rigorous studies, either do no good, do no good and cause side effects, or do no good and increase the risk of heart attack by 50%. Be sure to read about the just published data and the reaction to it in an excellent write up by Gina Kolata in The New York Times.

How could this be? Well, what if the things everyone thinks are causes of heart disease in diabetics are really just other effects of the real cause? Or, and this is where it really gets scary, what if what everyone thinks is a cause in need of eradication is in fact part of the body's defense mechanism against the real cause? For a discussion about how obesity may be just such a protective mechanism see "One of the Scourges of Modern Life May Have Been Profoundly Misunderstood" in The Economist's Science and Technology section.

The takeaway from all this can be found in the first article. While these treatments seemed logical (and as noted in the article, at every meeting "some academic" would always be going on about how elevated blood sugar after a meal was dangerous and had to be lowered until eventually doctors had put thousands of people on these treatments) it turned out they were instead dangerous and ineffective. That'll always be the danger when we attempt to deduce solutions based on just the known variables of a complex and only partially understood system.

Many Parents of Children With ASD "Have Relied Upon Practitioners and Researchers Who Peddled Hope, Not Opinions Grounded in Science and Medicine"

Three more strikes against the claim that thimerosol causes autism spectrum disorders (ASD).

Mead v. Secretary of HHS

King v. Secretary of HHS

Dwyer v. Secretary of HHS (- the quote in the title can be found in this opinion)

Be sure to read the discussions of Petitioners' expert Sander Greenland and his opinions about the limits of epidemiology - at least on the "what it can't rule out" end of the spectrum. One wonders what his answer would have been if instead he'd been asked to formulate an "intervention policy" based on the same epidemiological studies - all of which show no association between thimerosol and ASD, and none of which show a "clearly regressive" subset of ASD. See "Epidemiologic Measures and Policy Formulations: Lessons From Potential Outcomes", by Sander Greenland, free online.

FDA Finds No Increased Risk of Femur Fractures in Women Taking Fosamax

The FDA has reported on its ongoing safety review of oral bisphophonates and atypical subtrochanteric femur fractures. "Recent news reports have raised the question about whether there is an increased risk of this type of fracture in patients with osteoporosis using these [Fosamax, Actonel, Boniva, Reclast] medications" "All available case reports and clinical trial data were requested. FDA's review of these data did not show an increase in this risk in women using these medications."

Coronary Heart Disease: Neither Degenerative Nor Man-Made?

In "On to a Fifth Age? How About We Finish the Second?" we discussed a JAMA editorial wherein Dr. Michael Gaziano asserted we may be entering a fifth age of the so-called epidemiologic transition. These transitions are claimed to be changes in the primary causes of morbidity and mortality and Dr. Gaziano opined that we are moving into an era in which obesity and inactivity will drive preventable illness. We discussed the origin of the idea of epidemiologic transitions and questioned  whether we'd ever finished the second age which would have required the conquest of infectious diseases.

The so-called third age was supposed to be the "age of degenerative and man-made diseases" but it keeps turning out that many illnesses thought to be due to wear and tear, lifestyle or pollutants actually have an infectious disease process at their core. Now there's growing evidence that coronary heart diseases (CHD) may in many cases have more to do with a number of infections, including influenza, than with lifestyle or the environment.

Here's a link to a letter published in the Reflections section of The Lancet: Infectious Diseases that nicely summarizes the pre-1970 thinking that pointed to infections as the cause of CHD, the subsequent predominating narrative of chronic diseases not being caused by infections, and the new evidence that chronic diseases are in fact often caused by previously undetected infectious processes: "Inflammation as the Cause of Coronary Heart Disease". And here's a link to a written debate about "this nascent field associating chronic diseases with infections" from 2002 with the author of the recent Lancet paper cited above: "Debate on the Paper by Maria Ines Reinert Azambuja & Bruce B. Duncan".

Given the enormous renewed interest in infections as a possible cause of chronic illness and the ease with which scientists can now find traces of bacterial, fungal and viral DNA (or RNA) at the scene of the suspected microbial crime it's fair to assume that we'll be seeing many more such stories in the future.

You Know Those Mass Screenings for Prostate Cancer? Nevermind.

According to the Houston Chronicle the American Cancer Society has finally come to grips with mounting evidence that indiscriminate screening for prostate cancer causes more harm than good thanks to (a) the inevitable morbidity resulting from needless biopsies and surgeries due to false positive tests; (b) the realization that an awful lot of people who consider themselves "cancer survivors" would never have known they had cancer but for the screening test as their cancers would have gone away on their own or would have grown so slowly that they'd have died of something else before the prostate cancer became threatening; and, (c) the unfortunate fact that early detection, despite what everybody has been led to believe, does not mean that aggressive cancers can be cured - it just means that we get to be treated for them, and worry about them, longer.

Here's a link to the new screening recommendations: "Revised Prostate Cancer Screening Guidelines: What Has -- and Hasn't -- Changed"

Also of interest may be the readers' comments over at the Chronicle and elsewhere. Predictably there are two dominant camps. One sees this change as a nefarious plot by Big Pharma and Big Medicine to prevent early detection so they can make more money by making people wait until they need more expensive medicines and surgeries. The other one sees the new guidelines as a nefarious plot by Big Government to save money by preventing early detection so it can save money on treatment and hasten the deaths of Americans thereby saving money on Social Security payments as the cherry on top. I've run across veniremen able to hold both views simultaneously. But that's a discussion for another day.

Loves Safranin, Hates People

The New York Times has an alarming article about the rise of gram-negative bacteria as the cause of a staggering amount of morbidity and mortality among hospital patients. The rogues' gallery of gram-negative "wretched beasties" includes Klebsiella pneumoniae, Legionella pneumophila, Pseudomonas aeruginosa, Escherichia coli, Enterobacter cloacae and Helicobacter pylori.

For more on the problem see: "How Can We Stem the Rising Tide of Multidrug-Resistant Gram-Negative Bacilli?" wherein you'll read: "... we know much less about how best to detect and prevent transmission of multidrug-resistent gram-negative bacilli. These organisms are relative newcomers to the healthcare-associated infections arena ..."

Does Estrogen + Progestin Increase the Risk of Lung Cancer?

According to "Lung Cancer and Hormone Replacement Therapy: Association in the Vitamins and Lifestyle Study" published in the Journal of Clinical Oncology, long term use of estrogen and progestin (> 10 yrs) produced a 48% increase in lung cancer and typically advanced disease at that. Interestingly, those taking estrogen alone experienced no increase in risk.

The hazard ratio was only 1.48 with a confidence interval from 1.03 to 2.12 meaning the reported result suggests at best a hypothesis for further testing and not a presumed causal relationship.

Possible Link Between Low Level Benzene Exposure and Birth Defects, Miscarriages

In this article in last month’s Envrionmental Health Perspectives, the authors reported an increased incidence of sperm aneuploidy (abnormal numbers of chromosomes) in men exposed at air levels close to the current 1ppm Permissible Exposure Level (PEL) recognized by EPA. The chromosomes studied have previously been associated with birth defects and miscarriages. This study is the first to associate sperm aneuploidy with levels of bezene exposure below the PEL. It may be pointed to as some indication that the current PEL in the United States insufficiently protective against reproductive harm.

Household Exposures to PAH from Coal Tar Driveway and Parking Lot Sealants

In a paper published last month in Environmental Science and Technology, the authors measured polycyclic aromatic hydrocarbons (PAH) in settled house dust (SHD) in 23 apartments in Austin, TX. Half of the apartments had parking lots sealed with coal tar based parking lot sealcoat and half of them did not. The authors report 530 times more PAH in SHD in the apartments with parking lots sealed with coal tar based sealants than in the other apartments. These findings suggest that coal tar based sealants may play a significant role in household exposures to PAH, notwithstanding the fact that PAH are ubiquitous in urban environments.

The High Cost of Infections Acquired in the Hospital

According to "Clinical and Economic Outcomes Attributable to Health Care-Associated Sepsis and Pneumonia" just published in Archives of Internal Medicine just two types of preventable hospital acquired infections, sepsis and pneumonia, killed an estimated 48,000 Americans in 2006 and added more than $8 billion to the cost of care. The article details the alarming number of infections due to simple lapses in sanitation, from poor hand washing habits to dirty surgical devices, and the spread of so-called "superbugs" which have acquired immunity to common antibiotics.

Another Meta-analysis of Benzene and NHL Studies Finds No Independent Association

In "Benzene Exposure and Non-Hodgkins Lymphoma: a Meta-Analysis of Epidemiologic Studies" published in the Journal of Occupational and Environmental Medicine the authors looked at 8 cohort and 14 case-control studies of benzene-exposed workers and their risk of developing non-Hodgkins lymphoma (NHL). Whether in the aggregate or looking just at those with the highest reported exposures the summary relative risk estimate remained almost exactly one. The authors conclude that these studies at least do not demonstrate that benzene is a risk factor for NHL

Determining Causation After an E. Coli Outbreak

In a case involving lots of infected meat all from the same plant and all infected with the same pathogen cultured from the victims and known (thanks to the deployment of Koch's postulates) to cause the illness in question, well, tracing the source isn't too hard. But that's not how it usually goes.

After an outbreak and after all the food has been tossed and the food areas washed down with Clorox how could you possibly find the source? Thanks to the robust sales data kept by more and more businesses it may be possible to construct a "virtual cohort" and thereafter through the usual epidemiological techniques trace an outbreak all the way back to "cooked beef topside" at a single delicatessen. The method is discussed in "The Use of a New Virtual Cohort Study Design to Investigate an Outbreak of E. coli O157 Linked to a Supermarket Delicatessen".

Six new Stevens-Johnson Syndrome Papers in a Month

Among the conclusions: mycoplasma pneumoniae causes Stevens-Johnson syndrome (SJS) but symptoms were less severe than those cases produced by drugs and biopsy specimens were distinguishable; treatments for vision-related pathologies are available if administered timely; NSAIDs are only weakly associated with SJS and perhaps not at all; the percentage of total body surface affected by SJS/TEN (TEN is toxic epidermal necrolysis) is the best predictor of whether a potentially deadly blood stream infection will occur thus informing and the types of bacteria typically responsible may guide clinicians regarding antibiotic therapy; and, Tamiflu, SJS and you.

Does Hormone Replacement Therapy Prevent Colon Cancer?

A study of 57,000 California teachers found that those on hormone replacement therapy (HRT) had a 36% lower risk of colon cancer than women who never used HRT. The study followed California teachers for a over a decade. The study, "Menopausal Hormone Therapy Use and Risk of Invasive Colon Cancer: The California Teachers Study" is published this month in the American Journal of Epidemiology.

The effect, and by the way a 36% decrease in one of the leading causes of cancer is a pretty big deal, held whether the women were active or not, obese or not, taking NSAIDs or not, consumers of alcohol or not and was especially strong among women with a family history of colorectal cancer. Interestingly, the protective effect disappeared after 15 years.

So what should a perimenopausal/postmenopausal woman do? Is it just a matter of picking her poison? I've no idea. But until someone figures out how cancer really works it looks like we're stuck in a "damned if you do, damned if you don't" dilemma.

Soft Drinks and Pancreatic Cancer: A Plausible Link?

Late last night the press began reporting on a press release from the authors of "Soft Drink and Juice Consumption and Risk of Pancreatic Cancer: The Singapore Chinese Health Study". By this afternoon the claim that drinking as few as two soft drinks a week may nearly double the risk of getting one of those "go get your affairs in order" cancers had appeared on nearly every major news site. So what to make of it?

First it's probably wise to read up on the Singapore Chinese Health Study. How was dose (number of soft drinks per week) determined? It looks as though it was done via questionnaire rather than any objective measurements. And how many cases were there? 140; which isn't many thereby pushing the reported hazard ratio of 1.87 to the "slight" end of the range of the scale used to weigh evidence when making causal inferences.

Second let's look at the other claims on which the authors rely to support their causal hypothesis: specifically, a) Increased sugar intake may stimulate tumor growth through effects of insulin; and b) Pancreatic cancer rates increased nearly twofold over the past several decades.

As we reported recently there is indeed evidence tending to show that chronic moderate to high levels of glucose in the blood are associated with an increased risk of pancreatic cancer. Sounds supportive, right? Only if drinking sweetened soda chronically elevates blood sugar levels. And here's an article that says that it doesn't - not in lean men, not in obese men and not even in type 2 diabetic men; and not even at 75 grams of sucrose per day (seven times the purported risk elevating level referenced in today's article under discussion). So maybe it's like the fat follies of the past few decades; eating fat makes you fat, right? Wrong. Similarly, maybe the association between consuming sugar and hyperglycemia isn't so simple. And isn't that doubt furthered by the finding that sugary fruit drinks were NOT associated with pancreatic cancer?

Finally, here's the SEER Stat Fact Sheet for cancer of the pancreas. As noted there are lots of ways to look at trends but from the SEER perspective the upward trend in pancreatic cancer incidence and mortality is quite recent and looks to be driven by an increase among women. Doesn't that suggest smoking? It is after all the most consistently identified causative factor in cancer of the pancreas.

New Meta-Analysis Finds No Causal Association Between Benzene and NHL

A meta-analysis of 22 studies found no evidence of a link between benzene and non-Hodgkin's lymphoma (NHL) even at the highest levels of benzene exposures. Sliced and diced three different ways the data all revealed a summary relative risk estimate of almost exactly one every single time.

The study is titled "Benzene Exposure and Non-Hodkin Lymphoma: A Meta-Analysis of Epidemiologic Studies" and will be published in February's Journal of Occupational and Environmental Medicine.

Does Trichloroethylene Cause Parkinson's?

"I think people will really move on this as quickly as possible now" said Dr. Samuel Goldman of the Parkinson's Institute in Sunnyvale, CA according to the LATimes. Goldman was interviewed about a soon to be released study indicating that while other industrial solvents like xylene, toluene and n-hexane were not associated with Parkinson's, trichloroethylene (TCE) was strongly and significantly associated with the malady.

The results are to be presented in April at a meeting of the American Academy of Neurology and so there's not much to assess at the moment save the large grain of salt to be taken with any science that makes its way into the popular media before it lands in a journal. Nevertheless, the finding, if borne out by subsequent population studies, would be significant. Many substances have been blamed for Parkinson's but none with the 5.5 relative risk reported in this, well, press release.

Speaking of Saturated Fat

Yesterday we reported on a new paper in JAMA counseling caution in the face of activist efforts to impose universal sodium salt reduction on the American populace. We reminded our readers that once upon a time activists demonized saturated fats and stampeded the country towards greater consumption of trans fats. Now there's a new study looking at the health outcomes of 384,000 Americans to see whether those who consumed lots of saturated fats were more likely to develop heart disease than those who avoided saturated fats.

Reuter's Health reports that the study,published in the American Journal of Clinical Nutrition, involved the combination and assessment of 21 large studies of people, their diets with regard to the consumption of saturated fat and whether or not they developed heart disease. It turns out there's no demonstrable association between saturated fat and your risk of heart disease. Eat a lot or eat a little, your risk remains the same. Somewhere my late great grandmother, who lived to 102 and who made the best pies and pastries thanks in part to lard, is smiling.

Does Chronic Elevated Blood Sugar Cause Cancer?

It very well might, according to "Diabetes Mellitus Type 2 - An Independant Risk Factor for Cancer?" More alarming is the finding that significant risk of cancer is incurred even at blood glucose levels simply on the high side of "normal". A doubling of the risk for the following cancers has been found among those with type 2 diabetes: colorectal, breast, endometrial, renal, liver and pancreatic.

On to a Fifth Age? How About We Finish the Second?

In a 1971 paper that profoundly influenced how scientists and policy makers approached public health issues Abdel Omran set out his theory of "The Epidemiologic Transition". He hypothesized that societies went through three different ages, or phases, that defined their experience with regard to mortality and life expectancy. In the first, the "age of pestilence and famine", life expectancy is low and episodes of widespread death are common. In the second, the "age of receding pandemics", infectious diseases are overcome and life expectancy increases dramatically. Finally, in the third, the "age of degenerative and man-made diseases", diseases of aging and self-inflicted suffering becomes the predominant determinant of mortality. Eventually others, noting the dramatic increase in life expectancy due to the rapid decline in deaths due to heart attack and stroke, posited a fourth age; essentially the same as the original third age but with cardiovascular disease removed from the "degenerative disease" category.

Now in an editorial in this month's JAMA  Dr. Michael Gaziano asserts that we may be entering a fifth phase, or age, of the epidemiologic transition. We are now, he writes, entering the "age of obesity and inactivity" in which ailments due to gluttony and sloth predominate on death certificates. The editorial references two new articles in the same issue purporting to show Americans are fat and getting fatter; especially the children.

But wait a minute. The age of man-made diseases barely materialized. Certainly there have been many many cases of people suffering terribly as a result of some man-made health hazard. Look no further than the cases of mesothelioma among the men who served aboard amosite laden Navy ships. And smoking continues to exact its terrible toll. Yet if you throw all the deaths due to occupational diseases and every last lung cancer/COPD death into the same category you can't get to 10% using worst case estimates. More sober estimates put the percentage of deaths due to man-made diseases at considerably less than one. Nevertheless, this powerful meme - that most of our woes are self-inflicted and due to some failure to live in a natural way - still propels not only mass tort litigation but also much scientific and political thinking.

However, there's more than just AIDS to demonstrate that we never really saw the "disappearance" of infectious diseases. Go to www.pubmed.gov/ and do some searches on helicobacter pylori and humanpapilloma virus and you'll see just how many cancers are now being attributed to just these two organisms. Investigate mollicutes and you'll find that all sorts of microbes are suddenly being found associated with disease and they're only now being found because the technology to identify them is only now being refined.

Finally, remember to read the fascinating journey of Barry Marshall and Robin Warren from authors of an abstract rejected as one of the year's worst to winners of the Nobel Prize in Medicine for the very same work. In the end, the view, supported by the work of one of the world's preeminent public health researchers, that peptic ulcers were caused by that most modern of man-made insults, stress, only gave way to the understanding that the cause was in fact a bacteria when the evidence was irrefutable.

A Critique: Recent Epi Studies of Motor Skills and Manganese

In "Risk Assessment of an Essential Element: Manganese" Annette Santamaria and Sandra Sulsky of ENVIRON critically review recent epidemiological literature associating a variety of abnormal psychometrics with relatively low levels of manganese exposure.  

The authors conclude that the available epidemiological data is generally flawed and unreliable at least for the purpose of doing risk assessment. Furthermore, they demonstrate that some exposure levels claimed to pose a risk of neurobehavioral injury produce effective doses well below the amount of manganese recommended in a healthy diet; they also elaborate on the adverse health effects of manganese deficiency. Santamaria and Sulsky conclude by suggesting that more accurate and defensible risk assessments for manganese will have to come from objective data such as the determination of manganese dose via inhalation and the subsequent development of physiologically based pharmacokinetic models to predict the consequences of exposure at various levels and by various routes.

An Explanation for the Quebec/South Carolina Chrysotile Conundrum?

Why is there such a disparity between health outcomes of the workers who mined chrysotile and the textile workers who used it? In "Comparing Milled Fiber, Quebec Ore, and Textile Factory Dust: Has Another Piece of the Asbestos Puzzle Fallen into Place?" Berman reports that the dusts encountered at the facilities vary in two important ways. First, whether assessed by PCM or TEM, samples revealed that mine and mill dusts contain only 67% asbestos. On the other hand, typical feedstock for the textile facility was 100% asbestos. Second, the feedstock asbestos fibers were significantly longer than those typically encountered by workers exposed to mined or even refined asbestos.

Thus, for a given quantity of dust textile workers would have an approximately 50% greater exposure to asbestos and that exposure, on a fiber per fiber basis, would be significantly riskier. The author also suggests a method for reconstructing past exposures which would more accurately estimate the risk posed by each of those exposures.

Does Education Cause Autism?

If a strong and consistent association between autism and a single chemical or vaccine were found in ten separate clusters around California you'd expect to read about it; and you'd expect the researchers to infer that the chemical or vaccine was, in fact, the cause of autism. But what if the only strong and consistent association wasn't between autism and exposure to some substance but was instead between autism and something complex, like "high parental education"? Well, you'd probably expect a lot of frustration; and you'd be pretty sure autism wouldn't be laid at the feet of higher education.

So what should we make of "Geographic Distribution of Autism in California: a Retrospective Birth Cohort Analysis" ? Why would the children of parents with college degrees be at a 400% increased risk of autism? Why aren't chemicals to blame?

The authors concede in interviews with Scientific American that the results tend to undermine claims that industrial pollution causes autism. Yet rather than explore the possible reasons why college-educated parents might be significantly more likely to have autistic children the authors speculate that their results may be biased. Perhaps, they suggest, the less educated aren't educated enough to take their autistic children in to be diagnosed. Or maybe, they wondered, the educated are much more likely to use some unknown household chemical which does, in fact, cause autism. Scientific American hastens to add that there's an unpublished study that shows a doubling of the risk of autism in mothers who use pet flea shampoos.

But why point to a much smaller effect allegedly to be found in an unpublished paper when a much larger effect published in a peer reviewed journal is in front of you?

First, people tend to prefer simple answers with readily implemented solutions, the blame for which, and the costs of which, get imposed on someone else. 

Second, epidemiology, at least as practised in the West, tends to be reductionist. The triumph of linking tobacco to smoking led many epidemiologists to believe, or at least to want to believe, that indeed there are simple solutions to big problems and that silver bullets are out there just waiting to be found.

Finally, it's becoming increasingly clear that all sorts of ailments, from cancer to obesity, are highly complex and result in large part, from lifestyle, cultural and economic choices. Take breast cancer for example. One of the very best ways to reduce the risk of breast cancer is for a woman to bear children and to begin doing so in her late teens or early 20s. Yet no one wants to say that putting off children to get a degree and start a career is a cause of breast cancer.

And so, for now, the search is still on for the usual suspects.

 

 

Don't Pitch the Water Softener

Have you been worrying that your water softener is significantly increasing your risk of dying from a heart attack? I didn't think so. But just because you haven't been feeling vulnerable around your water softener doesn't mean the WHO hasn't been fretting for you.

Thanks to epidemiological studies going back a decade or more (e.g. "Magnesium and Calcium in Drinking Water and Death from Acute Myocardial Infarction in Women") a worry arose that we were killing ourselves by eliminating the minerals naturally found in most drinking water. Yet subsequent studies have failed to confirm the finding including the just published "Effect of water hardness on cardiovascular mortality: an ecological time series approach". So what gives?

Well, what gives is that most of what gets published in peer reviewed journals is probably false; and when it comes to causal inferences drawn from epidemiological studies "the apparently indiscriminate indentification of particular aspects of daily life as dangerous to health" is, as witty programmers say, a feature, not a bug.

 

Is Your Drinking Water Safe? If Not, Why Not?

The New York Times has run an extensive article about the nation's drinking water claiming that our public water is contaminated with thousands of chemicals, hundreds of which are "associated with a risk of cancer". It even has a link to the articles on which the claims are based. Unfortunately, or fortunately depending on your point of view, the evidence cited for the proposition that tap water is putting the citizenry at risk of cancer is pretty thin if that's all there is.

For example, under "Studies Regarding Illnesses and Drinking Water" (of which there are only eight) the only one to make a broad claim of tap water carcinogenesis is a 28 year old study titled "Cancer and Drinking Water in Louisiana: Colon and Rectum" which used the 1970 census to compare 692 rectal cancer deaths from 1969-1975 by where along the Mississippi River they and controls got their drinking water. The authors noted a small increase in risk as the source got closer to the Gulf of Mexico and suggested that the finding may have something to do with an increasing concentration of industries along the river as it approaches the Gulf. More importantly they wondered whether by-products of chlorination might have something to do with the finding.

The link suggests just four other papers that have cited the study and of those only two studied drinking water. Of the two, the first is a Canadian paper from 2000 (the first true Y2K victim I've ever run across - note: "Received 1999 Accepted 1900") which found no association with rectal cancer but a small one for colon cancer among males who drank "chlorinated surface water for 35-40 years".

Following the papers that cited the Canadian paper you quickly find another drinking water paper that finds a small protective effect for all leukemias combined, a large protective effect for chronic lymphocytic leukemia and a small but significant association with chronic myeloid leukemia.

The other papers are similarly all over the place and there appears to be no consensus that U.S. drinking water is a cause for concern about cancer from the perspective of chemical contaminants.

On the other hand, there's a growing body of literature associating drinking water contaminated by microorganisms with cancer. There's a new one discussing the waterborne transmission of helicobacter pylori to be published in next month's Journal of Water and Health and then there's this alarmingly titled paper in the same journal: "Free-living amoebae, Legionella and Mycobacterium in tap water supplied by a municipal drinking water utility in the USA

It's unclear why the NYTimes focused on trace levels of chemicals as a cause for concern when there does appear to be something to be worried about when it comes to bugs in our water.

Cancer: Annual Report to the Nation

There was more good news last week about Americans' incidence and rates of death due to cancer. The rates of new cases and the rates of death from all cancers combined are dropping and dropping steadily. For example, cancer deaths for all forms of cancer collectively dropped 1.6% per year for each of the last six years for which data is available (2001 - 2006). The progress being made against colorectal cancer was highlighted in the report.

While men continue to have higher rates than women their decline in rates was also higher. Decreases in deaths and incidence for lung, prostate and colorectal cancer drove the decline for men.

Among women, breast and colorectal rates declines lead the overall improvement.

The news was not uniformly good. In men, rates increased for myeloma, kidney, liver and esophageal cancer as well as multiple myeloma and leukemia. In women the expected increase in lung cancer rates was found along with increases for thyroid, pancreatic, bladder, melanoma and kidney cancers as well as for non-Hodgkin's lymphoma and leukemia.

The NIH press release can be found here.

IARC Assesses Evidence for Carcinogenicity and Genotoxicity of Formaldehyde

IARC has determined that there is strong evidence that formaldehyde is the cause of nasopharyngeal cancer and moderate evidence that it causes leukemia, especially of the myeloid variety. Here's the ungated table from "A review of human carcinogens - Part F: Chemical agents and related occupations".

Formaldehyde: A Cause of Leukemia?

In a new article in the Journal of the National Cancer Institute an association between embalming, as a proxy for exposure to formaldehyde, and myeloid leukemia was found to be strong, significant and dose dependent. In coming days we'll be discussing the controversial claim of a causal link between leukemia and formaldehyde and IARC's recent musings about it.

The article is Mortality From Lymphohematopoietic Malignancies and Brain Cancer Among Embalmers Exposed to Formaldehyde .

Interstitial Fibrosis Among Farmworkers Due to Agricultural Dust

A series of lungs from California farm workers who died in accidents or from illness were examined for evidence of pneumoconiosis.   Despite being young these Hispanic farm workers had significantly more evidence of interstitial fibrosis than did non-farm workers.  A variety of analytical techniques demonstrated that those with pneumoconiosis had significant exposure to crystalline silica and aluminum silicate.

The article is "Pneumoconiosis from agricultural dust exposure among young California farmworkers".  As for the source of mineral dust exposure the authors note "most agricultural soils are composed largely of silicate materials (e.g., feldspars, mica, clay minerals) and crystalline silica (CSi) (quartz)".

Are Big Punitive Awards in HRT Cases Justified?

Law.com is reporting that Philadelphia juries have awarded a total of $103 million in punitive damages alone to two women in separate breast cancer product liability trials. The women claimed that hormone replacement therapy (HRT) was responsible for their subsequent development of breast cancer.

In light of the recent controversy over the use of Bayesian decision-making approaches to mammography and Pap testing in which probabilities of outcomes are estimated and benefits are then weighed against costs (including other bad outcomes) I thought it might be of interest to see if such an approach had been applied to HRT. Sure enough, "Bayesian Meta-analysis of Hormone Therapy and Mortality in Younger Postmenopausal Women" was just published in The American Journal of Medicine.

So what does it show? It shows that across a number of randomized controlled trials of HRT in postmenopausal women under 60 those women had a reduced overall mortality compared to those postmenopausal women under 60 who weren't on HRT.

As is often the case in these modern times science does not yield a cure but does allow one to pick one's poison as it were; not to avoid death but to influence the odds of whether you die of stroke instead of breast cancer.

O-toluidine and Aniline Again Associated with Bladder Cancer

A recently published study in Occupational and Environmental Medicine reanalyzed the results of a 1991 investigation conducted by NIOSH of 1749 workers potentially exposed to o-toluidine and aniline at a chemical manufacturing plant and the incidence of bladder cancer. In the reanalysis, the investigators used updated exposure categories based on additional information ascertained through plant visits, employee interviews, documents, and answers provided by company and union officials on specific questions. Increased risks of bladder cancer were observed the longer the exposed employee worked at the facility and the longer the latency period from the date the employee initially worked in a department in which exposure occurred.

The investigators concluded that the findings were comparable to the results reported in 1991 by NIOSH that workers at the plant have an increased risk of bladder cancer.
 

Sulfonamides and Nitrofurantoin Associated with Birth Defects

The New York Times is reporting on this study by Crider KS et al. in November’s Archives of Pediatrics and Adolescent Medicine which appears to demonstrate an association between certain antibiotics taken for urinary tract infections (used just prior to pregnancy or during the first trimester of gestation) and several common birth defects.

The antibiotics associated with the most pronounced increases in birth defects were  nitrofurantoin and the sulfonamides. Caution is in order when interpreting these results. First, it appears that the exposures were extrapolated from memory rather than measured quantitatively. Questionnaire epidemiology is notoriously prone to error. Second, there’s the question of where does the arrow of causal suspicion point. Maybe it’s the infection that’s responsible for the birth defects. Finally, this particular cohort would not appear to be a cross section of pregnant American women but instead is drawn from a narrower slice and is significantly homogeneous. Social structures and interactions are increasingly recognized as important independent causes of certain morbidities.
 

Good News on Styrene

A review of the studies of workers exposed to styrene found no evidence of a causal link between styrene exposure and cancer. The literature contains numerous studies of workers with exposures to styrene during its manufacture or use in synthetic rubber or plastics production. Though a couple of studies have found an increase in either non-Hodgkins lymphoma or esophageal cancer many others did not and no dose response was seen in the outliers. The authors conclude that the available epidemiologic literature does not support a causal inference for styrene and any form of cancer.

The paper, published in the October issue of Journal of Occupational and Environmental Medicine is titled “Epidemiologic Studies of Styrene and Cancer: A Review of the Literature” and was authored by Boffetta P, Adami HO, Cole P, Trichopoulos D and Mandel JS.

Mortality Among 127,266 Petroleum Company Employees

The Journal of Occupational and Environmental Medicine is publishing in its October issue the results of an epidemiological study of 127,266 men employed in the U.S. by a major petroleum company with at least one day of employment from 1979 through 2000. Standardized mortality ratios were calculated for 94 causes of death. Overall these workers had risks below, and often well below, those of typical American males of the same age. However, the mesothelioma rate was triple that expected (CI=1.15 – 1.90) but only for men hired before 1960. Overall, the rate for acute non-lymphocytic leukemia (including AML) was essentially as expected though a subgroup in the chemicals portion of the business did show a statistically significant increase. Interestingly, there was a statistically significant increase in mortality from melanoma among truck drivers.

As is typical of refinery worker studies, these employees fared quite well as a group. Whether that’s because, as a plaintiff attorney friend of mine speculates, the oil companies are especially good at spotting people who’ll be less likely to die of say stomach cancer or whether it’s because these workers tend to be better paid, have better access to health care and so tend to live healthier lives remains to be seen.

The article is titled “Mortality Patterns and Trends Among 127,266 U.S.-Based Men in a Petroleum Company: Update 1979-2000” and was authored by Huebner WW, Wojcik NC, Jorgensen G, Marcella SP and Nicolich MJ.

Trichloroethylene: A Risk Factor for Cancer?

US EPA has been working on a risk assessment of trichloroethylene (TCE) for some time now. Here’s a link to the EPA Issue Papers through 2005. Now a comprehensive review of the issues has been published in Critical Reviews in Toxicology. The article is entitled “Trichloroethylene risk assessment: A review and commentary” and it provides an excellent overview of the developing molecular biological and molecular epidemiological approach to causal attribution and risk; one we’re sure to see increasingly in asbestos, benzene and other mass tort litigation.

If a Lead, Must It Bleed?

Over at Mind Hacks there’s a fun discussion about the media’s propensity to take epidemiological data regarding say the risk of some dread disease, to note the uncertainty inherent in such data and to then conclude that the risk could actually be much higher. Why not also note that the uncertainty also means the risk could be much lower? Is it just a matter of hyping a story to sell more newspapers?

I think it may be something else. I think it may have more to do with the product of the news industry – a narrative that customers want to buy. Accordingly, when the news seems not to fit with the readers’ worldviews (e.g. a report about the disparate outcomes between children in daycare versus those who stay home with their mothers) you can count on the reporter to note that the surprising result may be overstated. Thus, it may not be a matter of reporters not understanding such data or of wanting to fan the flames of a story but rather to pound the day’s news into a narrative to the liking of its readership.

Next week in Munich, Germany - The 2009 Benzene Symposium

Here's a link to the program and abstract book: http://www.tum-benzenesymposium.de/AbstractBook_FINAL_082509.pdf

Given the current state of benzene litigation among the most interesting presentations will be those pertaining to (1) latency (whether attributable cases drop dramatically after 10 years or so); (2) disease endpoints (just AML, just certain subtypes of AML, or NHL, MM and even ALL); (3) molecular biology/epidemiology (biomarkers and pathways) for assessing past exposures; and, (4) risk assessment.

Details to follow.

Diagnosing Mesothelioma

Distinguishing a mesothelioma from a lung adenocarcinoma is critical in asbestos malignancies. Over time, diagnoses made on the basis of morphology and the presence of asbestos bodies gave way to immunohistochemistry. Immunohistochemical staining panels keep changing and debates often rage over whether say a positive calretinen, etc. and negative CEA, etc. is enough or whether they merely show for example an adenoma of mesothelial origin. Now there's a new paper out discussing the use of epigenetic (your genes aren't so deterministic after all) profiles. Its title is "Differentiation of lung adenocarcinoma, pleural mesothelioma, and nonmalignant pulmonary tissues using DNA methylation profiles" and you can buy a copy of it there.

Of course all these new methodologies raise the obvious question of "What happens when you try to draw causal inferences about a case diagnosed today from epidemiological research conducted at a time long before these diagnostic techniques existed?" Will these then be, at least with regard to the litigation, distinctions without a difference?