Tuesday, March 20, 2007

Hilden on Zyprexa

Julie Hilden has posted a Writ column on the Zyprexa document leak.  She comes out essentially where I do on the propriety of the conduct of the parties involved -- suggesting that they should have tried the legal methods first.  I think without making that effort, any attempt to argue that the leakers were engaging in civil disobedience is silly.

She raises reasonable questions about the value of the protective order in mass torts in the first place, which, as I've noted before (and will address in an in-progress law review article) really depends on how much you consider the litigation system to be a mechanism for public information distribution as opposed to the resolution of private disputes.  I think people can reasonably come down on either side on that issue; I would prefer that the information-forcing take place through regulatory means rather than litigation, which is an extremely imprecise tool for such things, but then again, there's plenty of reasons to doubt the FDA's ability to do information forcing well, at least today's FDA.

And she also assumes something that I still haven't seen to be clearly true -- that the documents in fact provide reliable information that was not available before.  (I still have not reviewed the actual documents.)  Most of the coverage suggests that the documents have all sorts of unpleasant behavior by employees (including most likely improper off-label marketing), but I haven't seen the sort of thing that actual physicians would want to rely upon in making prescribing decisions.  Nor have I seen anything that indicates that the leaked documents contained, for instance, data that should have been given to the FDA but was not.  I'm open to being wrong on both counts -- and will be reviewing the documents for the article -- but it's not self-evident at this point.

Regardless, it's worth a read.

(Apologies for incoherence - it's my daughter's spring break and her friend's here, so there's more noise than usual...)


Experts & Science, MDLs and Class Actions, Products Liability | Permalink

TrackBack URL for this entry:


Listed below are links to weblogs that reference Hilden on Zyprexa:


Your own assumption is about what it is that actual physicians rely on to make prescribing decisions. Studies in the medical literature show that medical decisions are influenced by pharmaceutical marketing activities (and why else would companies spend so much money on them). A comparison of the sales figures for Zyprexa and the number of people in the population with a diagnosis of schizophrenia or bipolar disorder suggests that "most likely off-label marketing" has been quite effective.

The legal question for me is how long it would have taken to get this information into the public domain by "legal means", bearing in mind that every new patient prescribed Zyprexa was (and is) apparently another patient potentially injured.

Posted by: Robyn | Mar 20, 2007 7:03:23 PM

That's a good point and one I should have addressed. The danger of posting with a playdate in the background.

Most of the criticism of the company coming out of the documents, however, has been addressing things like knowledge of risks at various times. Risk evaluation is not done, by regulators or by physicians, based on concerned e-mails among exmployees at the company. I take it that you agree that to the extent the documents suggest that employee x was concerned about side effect y on date z, a prescribing physician doesn't need to know that, especially if all data from trials and AERs has been properly disclosed. (There's a dispute about that in the litigation, of course.)

But yes, the off-label information is relevant, certainly to someone. Do all physicians in the world need to know it? Maybe, maybe not. That's the sort of balanced evaluation that a court could make in response to a challenge of confidentiality, permitting (just as an example) submission of the key documents to relevant regulators, rather than a flood of thousands of pages -- many of which, from press reports, have nothing to do with off-label marketing and everything to do with embarrassment -- to the press.

And as to the speed with which the court would have moved, I'm not sure how that qualifies as a legal question, but it certainly would be relevant to someone's determination as to whether civil disobedience was justified.

Posted by: Bill Childs | Mar 21, 2007 3:46:10 AM

Regulatory authorities may or may not have had all the data from trials and the sub-set of adverse events that get reported, but there is much less likelihood that the average prescribing physician has had access to all of that information (a) because s(he) has not had time to seek it out for every drug they prescribe and (b) because many trial results are not published - or made available to them in other ways. If employee x (concerned about side effect y on date z) was from the research arm of Lilly, then yes, every physician who prescribes the drug after that date has an interest, on behalf of their patients, in what came next. Drug companies have a duty to communicate information about risks as well as benefits, a duty not often fulfilled well for the former in their marketing.

The documents which look like they are evidence of off-label marketing are very relevant. Drugs are licensed for specific conditions because those are the ones for which the data showing efficacy reached statistical significance. Zyprexa is licensed in the US for shizophrenia and bipolar mania. However, it seems that Zyprexa sales representatives have been encouraging doctors to prescribe it in large numbers for the agitation seen with dementia, for depression and even, in young people, for "mood disturbances". Weight gain +/- hyperglycaemia is a serious side-effect, but there is room for argument that it is an acceptable risk, if properly monitored, in illnesses as severe as shizophrenia and bipolar mania. But Zyprexa has apparently been Lilly's fifth to top seller, so many, many people have been exposed to this serious risk. (And some have been injured and some will have died. We will probably never know the extent - how thoroughly are deaths of the demented elderly investigated?) Even worse, though - because Zyprexa has not been shown to be effective in anything but shizophrenia and bipolar mania, it is likely that in a great many cases the weight gain +/- hyperglycaemia +/- diabetes has been unnecessary.

And Egilman, Gottstein and Berenson are irresponsible?

Posted by: Robyn | Mar 22, 2007 7:32:22 AM

Your final sentence is a bit of a non sequitur. As I said up front, I was and remain open to being convinced that the documents contain relevant information, though until I look at the documents myself, I'm not. (That's not a rejection of your view of them, just caution.)

You identify the correct questions to ask about the data disclosure (and I am quite aware of the benefits and limitations of the data disclosed in labels and in marketing) and about off-label marketing. As I explicitly stated in the comment before yours, I agree that the information about off-label conduct is relevant -- to someone. The FDA has fairly broad powers to require remediation of off-label marketing; sometimes they even use that power, and it is likely to be a more measured response than results from wholesale leaking of documents that may cause people who should be on the drug to take themselves off of it. All that suggests that an attempt to comply with the protective order to get the information out legitimately would have been wise -- and would have potentially eliminated the distraction of arguing about their conduct.

But the question about their conduct -- one on which Hilden and I appear to agree, incidentally -- has to do with whether they should have attempted to comply with the protective order before distributing the documents themselves. I think the answer is yes. They assert that they did attempt to comply with the protective order. Hilden, I, and (most important) Judge Weinstein think they didn't succeed in doing so.

Posted by: Bill Childs | Mar 22, 2007 11:20:56 AM

Post a comment