Drugmakers Actually Make Lemons, Not Medicines?

The lemon, of course, is a metaphor and is not a pretty one. But Donald Light, a sociologist and professor of comparative health policy at the University of Medicine and Dentistry of New Jersey, contends that the spate of instances in which drugmakers hid or downplayed info about serious side effects - or overstated benefits - has transformed the market for medicines into one for lemons.

"The basic idea is simple but arresting: just as there are hidden dangers and flaws in used cars, so there are in drugs. And just as used car salesmen have an incentive to profit from not disclosing risks the consumer cannot see under the shiny exterior, so drug companies and their reps have an incentive to profit from not disclosing risks the physician and patient cannot see inside a shiny new pill. In a number of ways, however, drug markets differ from used car markets. One way is that bad drugs do not drive good drugs out of the market, as Akerlof famously predicted in the article that helped him win the Nobel Prize. Rather, they stay in, mixed with higher risk drugs," Light tells us, after presenting a paper on the subject at the American Sociological Association annual meeting last week.

According to his study, which is not yet published, independent reviewers found that about 85 percent of new drugs offer few if any new benefits. At the same time, though, prescription drugs are now a significant cause of death in the US due to toxic side effects or misuse. He notes that drugmakers spend "two to three times more on marketing than on research," which are dollars used to persuade docs to prescribe new drugs. But the docs, he maintains, may get misleading info and then pass along the same info to patients. In his view, this is a "two-tier market" for lemons.

And in his study, Light says drugmakers have taken advantage of this two-tier system information market and monopoly rights to create what can he calls "the risk proliferation syndrome," which consists of corporate, government, and medical practices that maximize the number of drugs put on the market with risks of adverse reactions and then maximize the number of people induced to take them.

Why is this so? He offers three explanations:

...drugmakers are in charge of testing; "firewalls of legal protection" behind which negative info can be hidden, and what he calls a "relatively low bar" for drug efficacy required for new drug approvals. He cited one study of 111 final applications for an approval in which 42 percent clinical trials lacked adequately randomization; 40 percent had flawed testing of dosages; 39 percent lacked evidence of clinical efficacy, and 49 percent raised concerns about serious adverse side effects. "The result is that drugs get approved without anyone being able to know how effective they really are or how much serious harm they will cause," according to Light.

UPDATE: PhRMA sends us this statement: "The notion that new medicines are ‘marginally innovative’ is insulting to patients whose lives are being extended or improved by these breakthrough therapies. The paper, ‘Pharmaceuticals: A Two-Tier Market …’ makes a number of other misstatements of fact, among them: - Disclosing medicines risk: The Food, Drug and Cosmetic Act requires that all significant risks be disclosed in medicines’ approved labeling – even if no causal connection has been proven between the medicine and the observed adverse events. These risks are derived from years of clinical testing, typically in thousands of patients, as well as in postmarket experience. To allege a lack of transparency about the risks of medicines is flatly incorrect and not grounded in the reality that the labeling for FDA-approved medical products typically bear dozens of pages of listings of side effects and warnings.

- Investment in R&D v. marketing spending: America’s biopharmaceutical research companies lead the world in discovering and developing medicines that allow patients to live longer, healthier and more productive lives. Our industry invested an estimated $65.3 billion in 2009 in discovering and developing new medicines. That R&D investment by America’s biopharmaceutical research companies is roughly double the overall annual budget for the National Institutes of Health. And, by contrast, the CBO reported that pharmaceutical manufacturers spent an estimated $20.5 billion on promotional activities in 2008. - 'Low bar’ for regulatory approval: For every 5,000 to 10,000 promising compounds tested, just five medicines will make it to clinical trials involving patients and, of those, only one will eventually receive FDA approval. Our member companies devote an average of 10 years to 15 years on complex research and development and spend up to $1.3 billion on R&D for each new medicine approved. In recent years, clinical trials have become even more complex and more time-consuming, according a Tufts Center for the Study of Drug Development study completed this May.

11 Comments

Thanks Ed,

Things that make you go hmmmmm.

Since our society practices Caveat Emptor and we are now in an economic downturn perhaps even moving into part 2, it seems that patients/consumers may be taking matters into their own hands. See:

http://www.wolterskluwerpharma.com/Press/PI%20release%20031610Final.pdf

If this trend continues, Pharma beware - brand loyalty once destroyed may never be recovered.

If pharma becomes the commodity ie. a good for which there is demand, but which is supplied without qualitative differentiation across a market; as it seems to want to become, buyers will eventually be forced to treat it as such. Consumers will refuse to pay the premium price for the "commodity" item. Not a good business model.

Aug 23, 2010 - 3:18pm

Ed-Is there a link to the article? I can't find it on the organization's Web site.

Hi Salient Point,

Thanks for writing in and, no, unfortunately, the study is not yet published. The only document available is a press release, although I did communicate with Don Light as well.

Best ed

Aug 23, 2010 - 7:35pm

A couple of comments/concerns: A. If drugmakers are making lemons, then would insurance company subrogation clauses make lemonade out of the drug lemons? What about false claims...you worry that drug lemon victim restitution is a slippery slope that could get even more slippery.

B. Would a pattern of nondisclosure and/or under-disclosure of drug risks be a potential SEC compliance issue- for example, would it violate FCPA Books records and internal controls...and would it be covered by the D & O insurer in the event of a lawsuit and/or investigation? What about claw backs?

Condor Aug 23, 2010 - 11:49pm

Who KNEW?!

Who knew Ed could channel "Gucci Mane" -- on behalf of Big Pharma!? [Seriously do watch the linked YouTube rap video (even though there is a commercial at the head of it) -- lemon errything.]

Color me. . . chagrined -- with lemon errything, too. Really, Pfizer to Gucci Mane -- in one. Brrrr! Outstanding.

Namaste, one and all!

Aug 24, 2010 - 8:21am

Ed- I'm curious about the actual paper because the accusations are serious & rather astonishing. Randomization, for instance, is a fairly simple process. I've read hundreds of papers on pivotal trials & only one, from about a dozen years ago, seemed to have suspect randomization. If the author truly feels that more than 2 out of 5 pivotal trials have flawed randomization, it will be important to see how he substantiates this.

Aug 24, 2010 - 9:25am

If you offer a given clinical study to a skewed sample of patients, then there is going to be bias in the randomization as well. For example, knowing that a drug has a higher risk of adverse events in women and then skewing the drug trial towards enrolling men, and presenting the results as representative of all patients.

Aug 24, 2010 - 9:55am

While I don't have this particular paper, Don Light has written a number of other pieces that are related. Many of these should be searchable via the usual engines.

Aug 24, 2010 - 10:40am

The elephant in the living room appears to be all of the Pharma sponsored clinical studies in which results are not reported out.

For example Duloxetine (Cymbalta). A search at the FDA clinical studies site using the search terms "Duloxetine" AND "Closed Studies" AND "Eli Lilly" as the sponsor returns 80 studies as being completed. But only 15 of those completed studies reported out results.

I contacted FDA to inquire about what Lilly's obligations to report out are. FDA's response - None. The sponsor can pick and chose which study results to report.

So there are 65(!) Cymbalta studies with unreleased results sitting in Lilly's vault. You could probably make a lot of lemonade with those.

And I'm sure Lilly's selective reporting practice is replicated by all the Pharma gamers...er...players...er...whatever...

Aug 24, 2010 - 3:25pm

Chris-

It's true that the inclusion/exclusion criteria will affect the patient demographics. At the same time, FDA will narrow the indication for approval based on these criteria, which is why applicants usually try to go as broad as possible for pivotal trials.

That said, these criteria have nothing to do with the randomization process per se, which is what is allegedly flawed.

Aug 24, 2010 - 3:45pm

In my example, I am not suggesting this bias is related to the inclusion/exclusion criteria. If the trial is technically open to all patients, but the investigators tend to propose inclusion in the trial only to males, then any randomization will still be biased and non-representative of the patient population overall. Another example would be a drug where sicker patients might not be offered the option of participating in a trial because of physician concern that the patient might get placebo. The problem in these situations is that the trial is maintained to be fully randomized and representative of the patient populations overall with the condition.