Donald Light, a professor of health policy, has given a paper at this year’s meeting of the American Sociological Association, in which he estimates that “85% of new drugs offer few if any new benefits while having the potential to cause serious harm due to toxicity or misuse.”
Reasonable people know that this is likely to be true, though chances are their romance with a particular anti-depressant or an ADHD thing for their kid is still ongoing.
************************************************
And how do you keep the love alive despite side effects and enormous cost?
In a beautifully written and very sad opinion piece in the New York Times, Allen Frances pleads with his psychiatry colleagues not to introduce a new category into the next edition of the profession’s influential diagnostic manual:
Suppose your spouse or child died two weeks ago and now you feel sad, take less interest and pleasure in things, have little appetite or energy, can’t sleep well and don’t feel like going to work. In the proposal for the D.S.M. 5, your condition would be diagnosed as a major depressive disorder.
[This represents] a wholesale medicalization of normal emotion, and it would result in the overdiagnosis and overtreatment of people who would do just fine if left alone to grieve with family and friends…
But here’s the real beauty of it:
Because almost everyone recovers from grief, given time and support, this treatment would undoubtedly have the highest placebo response rate in medical history. After recovering while taking a useless pill, people would assume it was the drug that made them better and would be reluctant to stop taking it. Consequently, many normal grievers would stay on a useless medication for the long haul, even though it would likely cause them more harm than good.
Of course if pharmaceutical companies relied only upon this sort of stupidity on our parts, they wouldn’t be the successful enterprises they are. As Light’s paper demonstrates, an entire apparatus of disorder definition, control of the scientific literature, compensation for professors willing to promote new drugs, constant fear-mongering advertising, off-label use, regulator swamping, etc., etc., presses these pills upon us.
Those universities and medical school professors who continue to do their bit to fuck us up should stop.
**************************
UPDATE: The Telegraph discusses the article.
UD has just printed out a copy of Light’s article and is reading it now.
***************************
ANOTHER UPDATE: Citing current examples of global pharma corruption is shooting fish in a barrel, but to give Light’s paper some immediate context:
… [The Department of Justice] is … interested in corrupt payments that may have influenced the reliability or integrity of data in clinical trials performed outside the US. A recent report by the Department of Health and Human Services found 80 per cent of marketing applications for drugs approved by the Food and Drug Administration in the US had relied on at least one foreign trial.
“Companies may find themselves facing critical legal issues if approval of products rested on the results of studies the DoJ deems corrupt,” Arnold & Porter said in an advisory letter to clients last month.
… [L]ast November, Lanny Breuer, head of the DoJ’s criminal division, announced that investigators would be focusing on international corruption in the pharmaceuticals industry for “years”.
**************************
SUPERNOVA: Light’s paper is exploding all over the web. , blogs, people are talking about it.
August 17th, 2010 at 1:45PM
“Constant fear-mongering,” yes. The age of terrorism. Repent or die.
Meanwhile, real medical research is swamped by the muck and real suffering people can’t get care.
And the nightmare is less and less air-conditioned.
August 17th, 2010 at 4:48PM
“The market for lemons” is a metaphor. It refers to George Akerlof’s paper in 1970 concerning buying used cars. This article was rejected by several journals before finally being published. Who wants to hear about used cars?
The idea is that people selling used cars know about the cars, but the buyer doesn’t. The buyer assumes that the seller knows something bad about the car and isn’t willing to pay as much. So, people with good used cars don’t want to sell them, because they won’t get what the car is worth. In lingo borrowed from Gresham’s law regarding money, the bad drives out the good.
It turns out that this idea applies in a wide variety of economic situations, which is why the paper is famous. I can’t see how it applies to the pharmaceutical industry though. Good pills aren’t being held back.
Maybe “the market for lemons” applies to online education?
August 17th, 2010 at 5:22PM
It’s possible to make the argument that pharmaceutical companies are being made to look bad simply because of the situation they are in.
1) Ioannidis has made the claim that most studies in medicine are false (http://bit.ly/92phkr). His argument is based on Bayesian statistics (beyond me). Depending upon the numbers plugged into his model, 50-85% of studies “significant to p<0.05" are false (or false positive). Economists think 40% of their studies are false. Some of the problem is due to "bias," which could mean that otherwise well-intended investigators would like the medicine to work against an illness.
2) Studies the drug industry does, at FDA direction, are geared to showing efficacy. Studies are not powered for side effects. You could do that-require that no medicine is released on the market unless the risk for liver damage is <1:100,000 for instance, but it's just not done now. Information about side effects typically accumulates AFTER the drug is on the market. Additionally, the FDA's post-marketing surveillance is cumbersome, and neither the drug company, doctors, nor patients want to hear about side effects for a drug that seems to work, and it's clear you'll have an information delay in realizing that a drug causes side effects. It's wise not to take a drug until it's been out on the market for at least a year.
August 17th, 2010 at 5:46PM
To your last point, Brad: From pharma’s point of view, it’s wise – nay, urgent – to get people to take a drug right away. If they wait a year, as you suggest, problems with the drug may well manifest themselves… From a marketing point of view, you’ve got to make a killing fast.
And of course ordinary people don’t have this information — Light talks a lot about who knows things (pharma) and who doesn’t know things (us) — about how long something’s been on the market. Ordinary people don’t know as much as you do (I think you told me you’re a doctor?) about experimental protocols, FDA customs, and side effects, so they don’t have access to the very good advice you’ve just given: Wait a year on all new drugs.
I think it’s worth unpacking your suggestion that studies get distorted because well-meaning investigators want the meds to work… Put aside the fact that a significant number of investigators have financial conflicts of interest, so that their wanting isn’t well-intended, but self-serving. Concentrate rather on the remarkable notion (which I’ve heard from many sources) that all of these highly trained scientists are so emotionally invested in their pills that they bias (unconsciously?) the results. Really? That’s a scarier notion than conflict of interest, which is at least straightforward, and posits a smart, self-interested person as opposed to a person so emotionally out of control that they can’t do science.
Sorry. This line has always seemed to me an excuse. At least I hope it’s an excuse.
August 17th, 2010 at 6:09PM
Brad: I’m currently reading the entire Light paper, and he does make a rather good case for the applicability of the market for lemons idea to pharma.
Certainly, for instance, good pills are being held back. Pharma prefers, as you probably know, issuing slightly altered pills and trying to convince us that they’re really better than the pill they very closely take after, to engaging in the intensive lengthy uncertain research that might produce a really good pill, which is to say a really new formulation, new molecular entities, etc.
Light begins the paper by characterizing pharmaceuticals as “a two-tier market of information asymmetry in which actors controlling each tier has strong monopoly rights. Such a market rewards supplier-induced demand and the production and sale of ‘lemons,’ drugs with hidden side effects.” The actors he has in mind are pharmaceutical corporations and physicians, both of whom know far more than consumers. Light argues for “reframing professionally controlled markets as markets for lemons, where sellers can use their superior knowledge, expertise and authority to persuade clients that they need services or products of questionable benefit and possible harm.”
The most direct persuasion comes, of course, from incessant drug advertising that scares the shit out of people.
August 17th, 2010 at 7:11PM
Hmm…
I was getting ready to drop one and decided what the hell, then the last two posts popped up.
As we use to say in the med chem business: methyl, ethyl, propyl, butyl, futile… This refers to tweaking the chemical structure of therapeutic agents. Of course clever patent lawyers have made this game a little harder by claiming everything under the sun, regardless of the fact that RB Woodward might not actually be able to synthesize the patent covered molecule. This is a fundamental problem with the patenting system, but alas that is another can of worms. (So, uh, you’ve claimed propyl but not isopropyl, so we’ll make a molecule containing an isopropyl, patent it, market it, profit…)
I think UD’s right about the incremental improvement approach. If we don’t change things TOO much, then we may not have a new entity that goes to hell in a hand basket. And of course in marketing we always equate new with improved, which isn’t always the case.
And God forbid if someone actually come up with a totally new drug entity. Who could afford to test it? This is one of the reasons why the entry of universities into the drug discovery business is humorous to anyone who actually knows about drug development.
And so it goes.
August 17th, 2010 at 8:54PM
UD,
When trying to assess the benefit from medical devices or techniques, it is critical to assess the evidence suggesting, in point of fact, that sick care services — whether drugs or otherwise — are in reality only a very small determinant of population health. There is excellent evidence for this phenomenon both historically and in the present.
This hardly means, of course, that we should not take care of sick people. However social, political, and economic structures are arrayed, people will get sick and we should care of them. But the general, widely accepted notion among both lay and professional that high technology innovation of any kind is critical to population health is simply not evidence-based, in my view.
This is why I grow increasingly tired of the micro-level analyses of the costs and benefits of any particular drug or device. To be sure, it is critical to have the Donald Lights of the world suggesting just how low the benefits of particular products are, but the general idea that access to even effective acute care services and products are simply not prime determinants of health and illness in populations enjoys significant evidentiary support, on my reading.
(Our collective unwillingness to engage this evidence, even if it is to disagree with it, suggests a great deal of important points about American society itself, not to mention the rampant behaviorial pharmacologism as to pills).
August 17th, 2010 at 10:01PM
But Daniel, if I understand you correctly, you should welcome work of Light’s kind. It can be seen as a crucial preface, if you like, to the larger point you want to make about our unthinking reliance on technological innovation as a measure of health…
August 18th, 2010 at 4:56AM
Two issues, quickly, with maybe more once I read the paper:
A commons problem–if everyone gets and takes the advice to “wait a year”, no new information is created to further guide us on taking the drug. The only way, in some cases, to generate information is to have the drug out in the population, under a much better post-release monitoring system than we currently have.
To paraphrase Mark Pauly, I think, anyone who wants to volunteer to only receive 1950s medical technology at 1950s prices, please raise your hand.
August 18th, 2010 at 8:17AM
Do not worry. Under the Brave New World of American Medicine, Madame Pelosi will make the decisions about who gets which drug for which diseases. Dropping some serious coin into the coffers of the right candidates and causes will help undoubtedly her make some of these tough calls.
August 18th, 2010 at 12:32PM
UD,
You’re right, and I do welcome Light’s work in this regard. I just want to note my view that the larger context regarding prime determinants of health and illness in society is in my view the critical issue, or at least the critical issue I happen to be most interested in!
As an aside, might you be willing to let me or any other interested readers know how you acquired a copy of the paper in question?
August 18th, 2010 at 12:48PM
Daniel: I’ve forwarded the paper to you. Let me know if you don’t get it.
August 18th, 2010 at 7:56PM
How eerie. I met Donald Light seven years ago in connection with some of my own research, and for some reason was thinking about him yesterday for the first time since then…five minutes before I logged on to get my daily UD fix and saw your post!
August 19th, 2010 at 5:33AM
Arudra: Very gratified to hear of your addiction to UD.