When Lilly Pays Out $800M, Where Does That Money Go?
[Judge] Kelly immediately imposed sentence, ordering [Lilly] to pay $615 million, including a criminal fine of $515 million and $100 million in forfeiture.Not for me to judge the criminal sentence. Next:
The charge against Eli Lilly was announced earlier this month simultaneously with a civil settlement in the amount of $800 million. The total monetary settlement totaled $1.415 billion.
What civil settlement?
In the civil settlement agreement, Eli Lilly agreed to pay the United States approximately $438,171,543.58 to settle allegations that it caused invalid claims for payment for Zyprexa to be submitted to various government programs such as Medicaid, TRICARE, [etc] for unapproved off-label uses. Also, Eli Lilly agreed to pay various state Medicaid programs more than $361,828,456.42 to settle similar claims.
A.
First, an aside: why is it called "Zyprexa" in legal documents but "olanzapine" in medical journals? Is anyone fooled by this? This idiocy also extends to reporting about the drug: Psychiatric Times, same issue: FDA considers pediatric labeling for olanzapine; Lilly Off-Promotion of Zyprexa.
Richard when he gives you money, Dick when he doesn't, I guess.
B.
So $800M gets paid to the government because Lilly caused doctors to write Zyprexa off label.
Think about this. I'm not arguing the sentence. If the government simply wants to create a law that says off-label promotion is subject to a billion dollars in penalties, ok. Again, I'm no judge.
But under what logic can the marketing Zyprexa off- label be considered causing a false claim for payment? This isn't a clever legal trick: the False Claims Act is the explicit mechanism for policing off-label promotion.
Yet it's lunacy. The doctors wrote the script. Therefore it can never be a false claim, unless they're accusing the doctor of making the false claim, which they're not. If a doctor saw in a dream that Lipitor can calm an agitated demented patient and they try it, that's not a false claim. It's probably not even negligence.
The reason this matters is that we are in America, not Greece, or Spain, or Russia. Either we're upholding the law, or we're devising stratagems to go after whoever is currently politically unpopular. How you police something is more important than catching bad guys. This is why cops can't plant drugs on a suspect that they know is guilty. Right?
So if you're saying that Lilly caused doctors to unknowingly file a false claim, then you have to pick between one of these two choices:
a) either doctors were actually deluded by Lilly and thought that Zyprexa was FDA approved for "agitation in dementia," in which case it is very likely that their entire medical practice is one gigantic false claim; or
b) they knew that it was not approved for that, but were convinced by Lilly's marketing that it might actually work for that, so they prescribed it.
If you chose b), it should be immediately obvious that Lilly "caused" only Day 1 of the Zyprexa Experiment; everything else after that was the doctor's determination of whether it was working, worth it, etc.
Before anyone argues with me, understand what's at stake: physician autonomy. Anything that "causes" doctors to do something against their will or judgment; anything that "helps" physicians "practice better"-- all of these things presuppose that doctors are not able to figure things out for themselves. So then why have doctors in the first place? You may as well boot up a few Prescribatons and upload the Texas Algorithm 2.0 and let them medicate the consequences of the rest of the government's failed social policies.
Why resort to saying that off-label promotion is bad because it does something to doctors instead of saying it is bad in and of itself?
Oh: because when the government says a doctor was caused to do or not do something, what's soon to follow is the government telling a doctor to do or not do something.
Also interesting/horrifying, is this allegation:
Eli Lilly trained [its reps] to promote Zyprexa by focusing on symptoms, rather than FDA approved indications.
If we can just put aside our hatred of Big Pharma for one second, can anyone explain the difference between "symptoms," and an "indication" having no other objective characteristics except those symptoms? Oh, I hear the arguers clicking the caps lock. Be careful what you wish for. Would it have been better if doctors were "taught" to recognize "dementia related psychosis" as a schizophrenia-spectrum disorder? Oh, like you can tell them apart on MRI? Dementia praecox, after all.
I may be quick to point out the idiotic things doctors do; but none of those things, none of them are more idiotic than the things doctors are told to do, under the pretense of science (e.g. clinical guidelines) or the law.
C.
BTW:
The qui tam relators will receive $78 million from the federal share of the settlement amount.
Those would be the 6 or so former reps turned whistleblowers. No honor among thieves, or something.
Digg | 
del.icio.us | 
StumbleUpon | 
Reddit | 
Yahoo! | 
My favourites
March 16, 2009 9:13 PM | Posted by : | Reply
The "indication" verses "symptom" language is absurd. Reminds me of the "treat the symptoms" verses "treat the illness" dichotomy of alternative medicine. I don't know what it means to treat cancer verses the symptoms of cancer.
I'm in favor of evidential standards, but that's assuming people aren't over-stating the quality of the evidence. False certainties are dangerous.
Why did Lilly roll over? Did they offer a defense?
Score: 2 (2 votes cast)
March 16, 2009 11:21 PM | Posted by : | Reply
The federal government went after Lilly, because everybody knows big pharma is bad. And - more importantly - has deep pockets.
Individual doctors, despite their immense wealth, cannot cough up the required dollars to make such judicial jujitsu worth the effort.
We're going to see more of that in the future.
Score: 1 (3 votes cast)
March 17, 2009 9:26 AM | Posted by : | Reply
I think the biggest distinction here is that Lilly was marketing Zyprexa for uses that, while related to its certification, had not gone through the FDA. Regardless of our opinions about the FDA's efficacy or credibility, it was pretty certain that this practice would draw fire.
Score: 2 (2 votes cast)
March 17, 2009 9:56 AM | Posted, in reply to , by : | Reply
Seriously? If someone has a brain tumor, it may give them a headache. If you give them aspirin for the headache, you're not treating the tumor. That's the difference between treating the cause vs. treating the symptoms. "Alternative" medicine might also seek to find the cause of the tumor and treat that (i.e. lifestyle or environmental causes).
As for you, Mr. LP, I love your work, but I don't think this is what you make it out be (and I'm basing this entirely on what's posted here; I could be wrong and I'd appreciate more information). This:
So $800M gets paid to the government because Lilly caused doctors to write Zyprexa off label.
does not seem to follow from this:
In the civil settlement agreement, Eli Lilly agreed to pay the United States approximately $438,171,543.58 to settle allegations that it caused invalid claims for payment for Zyprexa to be submitted to various government programs such as Medicaid, TRICARE, [etc] for unapproved off-label uses. Also, Eli Lilly agreed to pay various state Medicaid programs more than $361,828,456.42 to settle similar claims.
If I'm reading this right, the problem was not that they caused doctors to write prescriptions for off-label uses.
"But under what logic can the marketing Zyprexa off- label be considered causing a false claim for payment?"
It can when the government will not pay for off-label uses. If they're marketing Zyprexa for off-label uses, with the understanding that they're going to be able to sell more of the drug because Medi-whatever will pay for it, then they're held accountable. I'm not saying this is right or wrong, but it doesn't appear to be about the doctor's writing prescriptions; it appears to be about the government then paying for those prescriptions.
Just to bring it back around to what I started talking about:
"If we can just put aside our hatred of Big Pharma for one second, can anyone explain the difference between "symptoms," and an "indication" having no other objective characteristics except those symptoms?"
While it doesn't apply to all psychological problems, here's an explanation: if someone had a traumatic experience that causes them to have angry, emotional outbursts over stupid things, you could give them a drug that would make them less emotional, or you could help them learn how to relate to their trauma better, so that it no longer causes these emotional outbursts. Obviously, not all psychological problems are mental, but that would be one way of explaining it. However, I don't think this is what they meant in that allegation, as they seem to conflate "indication" with "cause," which is not the same thing. Bipolar is not a cause; it's a label for a collection of symptoms that often appear together, which point towards a common cause, but it's not the cause itself.
Score: 5 (5 votes cast)
March 17, 2009 10:59 AM | Posted by : | Reply
The fact that you cannot reliably distinguish schizophrenia from dementia related psychosis on fMRI does not mean we should not respect the current conceptualizations of these disease constructs, which are solely defined by phenomenology, including the disease onset and course. When we lose this respect, we get a bunch of drug addicts/borderlines being switched from one "mood stabilizer" to another because they have "Bipolar." The State is of course totally hypocritical, because it actually prefers doctors to diagnose conditions that can -supposedly- be treated by medications (i.e. bipolar) so that more expensive psychosocial treatments (i.e. residential rehab) can be avoided. Drug companies capitalize on this bias and then go under when they take it too far.
In terms of Eli Lilly: I have no pity for them! Because of Eli Lilly's aggressive marketing to primary care physicians, thousands of patients needlessly developed morbid obesity and metabolic abnormalities, and God knows how many died as a result. They sat on this data and continued their marketing campaign fully knowing that their product was taking lives. This is the real criminal act here. Overall, I cannot understand why doctors allow drug reps "educate" them, when the literature is readily available. Too much facebook? Too many patients? What? If we are not going to put our act together, I think we only deserve more regulation and strict algorithms that will dictate us what to prescribe. "Have we no sense of decency, sirs, at long last? Have we left no sense of decency?"
Score: 2 (2 votes cast)
March 17, 2009 11:15 AM | Posted, in reply to , by : | Reply
No argument with your last paragraph; but again just because you know they're guilty doesn't mean we can create ways to prove them guilty. There is no need to bring a doctor's behavior into this case; by doing so, it encourages government to have greater oversight of doctors. By which I mean: fire them and replace them with Prescribatons.
Score: 4 (4 votes cast)
March 17, 2009 11:51 AM | Posted by : | Reply
"either doctors were actually deluded by Lilly and thought that Zyprexa was FDA approved for "agitation in dementia," in which case it is very likely that their entire medical practice is one gigantic false claim ...."
No, just one giant false claim to have the expertise to prescribe most everything. Reliance on outside (read pharma) resources to provide "information" about drugs like zyprexa is pandemic.
Score: 1 (1 votes cast)
March 17, 2009 5:18 PM | Posted by : | Reply
I believe the physician autonomy issue is a big part of this. The pharmaceutical company and the physician are not on equal footing. Sure, physicians are free to refuse to see pharma reps. But think about how challenging it is, currently, for a practicing physician to practice decently, clinically, with no dependence upon pharma.
Peer-reviewed journals, the FDA-approval process, professional conferences, thought leaders, and the CME requirment all seem reasonable. Professional. Even-handed.
But Big pharma has taken each of these avenues and used it as a marketing dimension. That doesn't mean a doctor HAS to prescribe as marketed, but it sure provides yet another influence. This is power. The doc does not have time to discover the empirical literature on each issue facing him or her. If he or she did have this time, the doc does not have the eudcational background to consume, much less digest, the information. If one meta-analysis uses a fixed-effects model, and another uses a random-effects model, how might that impct the resulting point estimate and confidence interval? And how might that translate to the decision whether to augment or not? Docs don't learn this in med school. Nethier do the drug reps, generally, but the drug reps are educated by their employer regarding how to interpret results. So they can reassure the doc regarding whether the result, fixed-effect model or not, says to augment or not.
When someone comes in with sleep complaints, a doc can either assess whether the person is about to go into foreclosure, or the doc can see whether the person meets criteria for being prescribed Seroquel for sleep.
Seroquel for sleep? Google it and you will find that, yes, Seroquel is being prescribed for sleep. If the research concerning side effects of Seroquel has been dishonest, then there is a power differential; Pharma can manipulate the physician.
Yes, you check the boxes and sure enough, on the little checklist with the logo at the top (or bottom, or wherever), it suggests "sleep disorder."
All the while, the person does not hav a sleep disorder. They have a financial management problem. But what harm is there in Rx Seroquel for sleep?
Peer-reviewed journals, the FDA-approval process, professional conferences, thought leaders, and the CME requirment. All say: no harm.
Pay no attention to the man behind the curtain.
Score: 2 (2 votes cast)
March 17, 2009 7:09 PM | Posted by : | Reply
I can't seem to find an email link, so this is entirely unrelated to the post.
How does one approach medical researchers as for requesting access to their data? I'm a professional econometrician by day, bayesian revolutionnaire by night particularly interested in mixing distributions right now (what happens to your effect sizes when you mix MDD and BP2 patients in a bag, for example), and failed to get in touch with the author of a paper on axis II correlates of bipolar disorder despite being standardly polite and attempting to establish credentials as a statistician.
I'm guessing there's something in the language or in failure to convey the impression of not being a complete crank (I did disclose my axis I diagnosis on honesty-is-the-best-policy grounds). The thing is, I don't have much of a publication record, and even what I can put together after negotiating on all the NDAs doesn't support "theoretical statistician" even by a prayer. (I like to compare myself to this physicist who made out big while working in a patent office, but that's my cranky bit.)
Score: 1 (1 votes cast)
March 18, 2009 10:03 AM | Posted by : | Reply
Diego - researchers don't share data too much, despite this general ethos that data should be shared with qualified people upon request, as part of the philosophy of science. I have asked for similar data - std dev's when only means were published - in order to relate my study sample to a similar, published sample.
What worked for me was name-dropping a collaborator who is well-known in the field. For my purposes, I actually wanted to get two different authors/research groups to provide additional data beyond that published. Both authors complied. So, it does happen.
Like you, I presented my role and my objectives, and was overly polite, etc.
I might suggest you team up with a local professor or researcher. when a request comes in with someone's academic title, and institutional affiliation, it is a decent "calling card."
If you have a decent idea that would be publication-worthy, then the academic could share in the glory of authorship.
Statistically, when you mix effect sizes from two similar pops, as you could guess, the point estimate of the efect size will end up being an average of the two contributing effect sizes, with adjustment according to whatever adjustment you use to accomodate the different sample sizes and different confidence intervals. Plus a hit to degrees of freedom, since any study subject was probably not free to end up in either study.
The main effect will be the on confidence interval. It will obviously get broader, which in the Bayesian analysis would just weaken the confidence interval on your predicted probability.
If you are an enthusiastic amateur on this topic, you might want to also consider a discriminant function analysis: trying to maximally distinguish groups (MDD versus BPD) on some set of measures (however you have bravely decided to measure axis 2 traits). SPSS and other stats packages will do it; you just need to read up enough to know what you are gonna point-and-click, and know how to interpret all those numbers that pop up in the output window. Not really too hard to learn or interpret - you will get classification data similar to when you run a logistic regression (how well does your weighting, or probabilities assigned to each predictor perform in placing cases in their actual category). The program will calculate a "centroid" for each of the two groups (MDD, BPD), and weight each predictor to maximally classify / reduce variance between each predictor and the "centroid."
Now that I explain it that way, it is like trying to maximize explanatory power / weighting of predictor variables of two linear models at once. which I guess it is.
Score: 0 (0 votes cast)
March 18, 2009 11:45 AM | Posted by : | Reply
I know linear discriminant analysis. I'm working on nonparametric, maximum entropy/Kullback-Leibler divergence approaches.
Score: 0 (0 votes cast)
March 18, 2009 5:56 PM | Posted by : | Reply
I agree with all of the comments made here- they are all valid. What it really comes down to is intent- did Lilly intend to defraud the system. Ultimately, it is up to a doctor to prescribe an appropriate medication. I read the full press release regarding this matter. The off label use was not only fraudulantly claimed through public insurance mechanisms but also was unapproved. That's why organisations such as the FDA exist- to guarantee that drugs that come to the market are safe, with full disclosure on the side effects. Without this, it is difficult for a clinician to make weigh the potential treatment benefits againsts its risks.
Ultimately, there are valid arguments on both sides of the fence. It's a hefty settlement and given the financial climate, I'm sure Lilly will be feeling the pain. Lilly have been pushing the dementia angle for a long time.
Score: 1 (1 votes cast)
March 19, 2009 1:25 AM | Posted by : | Reply
Off-label use is common, particularly with kids. Studies are done to support the off label use. But that doesn't always become an FDA indication, as those cost about a billion dollars.
If Medicaid and Medicare don't cover off-label use, the pediatricians can all retire now.
Score: 1 (1 votes cast)
March 19, 2009 1:28 AM | Posted by : | Reply
The tumor-headache example may seem straightforward. But the body doesn't draw a line between disease process and symptoms. It's a package deal.
Pain is part of cancer, and cancer is part of pain.
Score: 1 (1 votes cast)
March 19, 2009 1:43 AM | Posted by : | Reply
The drug reps who visit me drop off journal articles, some patient handouts, and some glossy stuff with graphs. They ask if I need any samples and go away after five minutes.
Occasionally someone new and eager wants to chat. I tell them they can't come back if they talk to me for more than five minutes.
The reps don't give lectures pimping the drugs. Doctors give talks on clinical issues and the drugs are mentioned in that context. So maybe the doctors should be sued for talking about off-label studies?
Perhaps Lilly tarted up Zyprexa in the glossy handouts. I never read them, so I can't say.
Score: 0 (0 votes cast)
March 19, 2009 3:11 PM | Posted by : | Reply
"why have doctors in the first place?"
Why weren't the doctors responsible for death and diabetes etc in their patients?
https://thelastpsychiatrist.com/2009/03/biology_is_destiny.html
"A student of history knows that history repeats; what is different is the people living it, and how they frame their present."
History is repeating itself , though not so bad this time in the numbers of the dead.
Why were only 23 out of thousands of psychiatrist charged with killing of 250,000 in Action T4 at the end of WW2, when it was hundreds of psychiatrists-doctors responsible.
Score: 0 (0 votes cast)
Comments