Medical Research and the Moral Dilemma of Pharmaceutical Industry Involvement

A medical school colleague laid out a moral dilemma yesterday that illustrates the lack of a simple answer to the question of whether involvement of industry in clinical research is good or bad. We are having a civil war in the medical research community, stimulated in part by Senator Grassley’s investigations, over whether we can or should be free of all industry contact. In some researchers’ eyes, contact with the for-profit sector is a conflict of interest that will inevitably corrupt the scientific and clinical enterprise. Others see industry involvement as a non-issue because they believe that scientists will report the facts objectively rather than skew them based on any money they might receive.

Here is my colleague’s not-as-simple-as-all-that dilemma. He has pilot data showing that a cheap, generic drug has a potentially life saving medical application. He now has a grant from the NIH (i.e., not industry) to study it in a major trial. A for-profit company has a copycat version of the generic drug that is virtually the same but is still on patent and is thus highly expensive. The question is whether he should study the generic version or the proprietary one.

If the experiment proves that the generic version of the drug is effective, he will get a very nice paper in a medical journal. And that will be the end of it. We have a million studies showing that clinical practice is minimally influenced by journal articles alone. In contrast, if he studies the proprietary drug, a profit making company will send out their armies of marketers to physicians and get many of them to use it, making a great deal of money for themselves in the process…but this will also save lives, because aggressive marketing and promotion such as the industry does has been proven to affect clinical practice dramatically.

What would you do and what would be your reasoning? I will post back here later what my colleague decided.

UPDATE — Provision of more information per comments so far: Studying both drugs is impossible, that would require a third study arm and a 50% budget increase, so a choice is ineluctable. If successful, the study would be a new use of both the generic and the proprietary version. The grant includes zero money for the scientists to disseminate the results.

SECOND UPDATE 21 JAN 2011 — My colleague decided to study the cheap drug with the expired patent.

Author: Keith Humphreys

Keith Humphreys is the Esther Ting Memorial Professor of Psychiatry at Stanford University and an Honorary Professor of Psychiatry at Kings College London. His research, teaching and writing have focused on addictive disorders, self-help organizations (e.g., breast cancer support groups, Alcoholics Anonymous), evaluation research methods, and public policy related to health care, mental illness, veterans, drugs, crime and correctional systems. Professor Humphreys' over 300 scholarly articles, monographs and books have been cited over thirteen thousand times by scientific colleagues. He is a regular contributor to Washington Post and has also written for the New York Times, Wall Street Journal, Washington Monthly, San Francisco Chronicle, The Guardian (UK), The Telegraph (UK), Times Higher Education (UK), Crossbow (UK) and other media outlets.

42 thoughts on “Medical Research and the Moral Dilemma of Pharmaceutical Industry Involvement”

  1. It looks like the classic clash of deontology versus consequentialism. Therefore, my choice depends on biorhythms and epicycles, as would that of any normally functioning human being.

  2. Almost total change of topic: my favorite example of this sort of thing is a study done in Australia a few years ago which showed that inhaling concentrated salt water was helpful for cystic fibrosis. (Note: study not done here. OF COURSE not done here. Although the inspiration was the observation that surfers with CF were doing better than non-surfers with CF, which is pretty location-specific…)

    My husband, who has cystic fibrosis and a lot of lab experience, has been making his own sterile saline in our kitchen at home — he can't face even his insurance company having to fork over $5 per vial, or however much they want, for pharmaceutical grade salt water.

    So now his doctors have been asking about his "recipe," because there's apparently A SHORTAGE of pharmaceutical saline of the correct type.

  3. No drug or medicine should be used in the United States until it has undergone a 20 year test. I would have no objection to having critical ill patients be free to take anything that they want.

    The whole nonsense of the drug industry defies any notion of fair, right, ethical or honest.

    I have mental health problems, multiple, interacting problems. I know what works best for me: smoking cannabis, smoking hash, smoking opium and taking LSD and mescaline. So I won't function in society? People, I'm not functioning now and relief is not just a pill away. Rehabilitation won't work and neither can I. We could cut the hell out of wasted money on under researched and dangerous drugs that help the arch criminals in the chemical industry hold up patient's health for pure greedy selfish self interest by simply legalizing drugs. I really don't care if Joe Biden (Ind – shill for Dupont Corp) starves when his Dupont Corp shares drop through the floor.

    I don't dare use the illegal substances, I have a very highly developed case of claustrophobia.

  4. Suppose his study shows the drug does not have the potentially life saving application. Will the army of marketers spread out to prevent its use?

  5. I'd say study the proprietary drug. Since patent terms are fairly short, the change in practice will happen, and then the drug will come off patent and become a cheap generic. (Short-term, there's more increase in cost and more general change in practice; longer-term, the cost increase goes away.)

  6. If the study is performed on the proprietary drug, and it has positive results, won't manufacturers of the generic drug also benefit? If so, opting to do the research on the proprietary drug seems to be a win-win.

  7. I'm not sure I understand the rules of this game.

    Is there no way to disseminate the result other than through a journal article? That strikes me as a self-imposed constraint.

    I'd say that, assuming a favorable result, your colleague has the obligation to make the outcome widely known. Whether that means working on the proprietary drug and then trying to piggy-back word about the generic on the company's marketing campaign, or doing something else I don't know. Won't the generic manufacturer put at least a little effort into publicity?

  8. I'd study both drugs. In the academic paper, I'd point out whatever differences the data suggested. If, as seems likely, the differences are minimal between the generic and the proprietary drug, I'd say so loudly. If the generic appears to be even statistically (i.e., not practically) better than the generic, I'd really say so. But that's just me.

  9. "We have a million studies showing that clinical practice is minimally influenced by journal articles alone." If doctors are only influenced by Big Pharma marketing and not at all by science, then the US health system is more broken than even I imagined. You really badly need a NICE. But I wonder, how does surgery advance?

    Study the generic. Then push the results through the relevant asociation of victims; every disease has one. Yup, the utilitarian calculus is omelettes and eggs.

  10. Academic detailing, in which a qualified expert visits physician offices to provide evidence-based information about therapeutics, is worth supporting in any version of health care reform. Industry detailing is an unsatisfactory substitute.

    HR 1859: Independent Drug Education and Outreach Act of 2009 would have provided funding for academic detailing; it was referred to the House subcommittee on Health and never got any further. But with prescription drug costs being a major part of out-of-control spending, there ought to be a strong push to reintroduce this bill.

    Then he could study the generic drug, publish the results, and have it introduced to physicians via an approach that they actually trust more than industry detailing.

  11. Color me equally puzzled by the conditions established to frame the debate. BY DEFINITION AND LAW, a "generic" product is equivalent to a proprietary one. There are some people who maintain that certain generic formulations are not quite the same, but the FDA has never seen a study that convincingly demonstrates this claim. I suspect that the missing ingredient is as follows: No manufacturer has ever sought FDA approval of the intended use being studied by the researcher, hence, neither the generic nor the branded drug can be marketed for that use. However, this will still be true after the research has been done unless a manufacturer seeks to expand the scope of what is referred to as an NDA — AND THIS WILL BE TRUE FOR BOTH THE GENERIC AND THE BRAND NAME MANUFACTURERS. THE GENERIC DRUG CANNOT MARKET THE DRUG FOR ANY BROADER APPLICATION THAN THE BRANDED DRUG CAN — UNLESS AND UNTIL ITS MANUFACTURER SEEKS THE REGULATORY EXPANSION.

    Okay, so this question cannot be answered without understanding the regulatory framework, and in our regulatory framework, it probably makes more sense to work with the proprietary drug, because the manufacturer is more likely to seek the NDA expansion than the generic maker.

    However, let me also say: the really pernicious impact of pharmaceutical money on research is that it sets the debate as one that ASSUMES a pharmacologic intervention is the answer. Hence, the cystic fibrosis example above — how many people are taking drugs when they might be able to get by with simple saline? Likewise, how many drugs are being funded for a condition like osteoporosis when research might show that making lifestyle or dieatary changes at a certain point in one's life is equally effective? Most acutely, how much research is being done to develop tests (like a pap smear) that would detect cancer at a stage that precedes by many years the stage that requires high expense tests (like CAT scans) or last ditch, super expensive intravenous drugs, compared to developing those super expensive intravenous drugs? Convenient, comparatively inexpensive diagnostic tools are what are needed — instead, pharmaceutical companies want to develop relatively expensive, sure to be less effective tools for treatment. Generic verses proprietary versions of a drug is barely an issue anymore.

  12. I am not sure this person should be thinking ahead so far. He hasn't proved anything yet and it's early to be worrying about publicity.

    If I were a scientist, seems like I'd want as much info as possible, so if it doesn't harm your study design, why not study both?

    I do think there are probably differences between generics and others, because anecdotal evidence from individuals about their bodies' reactions are valuable. Who knows which of these will work better, or at all?

  13. The ur-model of for-profit pharma is cheap-to-make, addictive drugs, which the user wants desperately — in other words, something like heroin, but failing that ideal, maybe aspirin or ex-lax. (Yes, shockingly, lots of people become addicted to aspirin and to laxatives — that's one factor in making them mass-produced, mass-advertised consumer goods.) Cheap cures are the last thing a for-profit pharma company will normally work on. Vaccines and antibiotics will tend to be neglected. Enormous amounts were spent developing chronic ulcer treatments, but no one in Big Pharma was involved in figuring out that one of the main causes of ulcers was a bacteria treatable with antibiotics.

    The economics of all this is not difficult to figure out, or subtle or mysterious. Just tragic and frustrating — like human nature.

    All the things Barbara said. But, the researcher should probably research the generic formulation. If he sees positive results, the for-profit company will be motivated to replicate the research and seek the NDA for its formulation (and, unfortunately, a patent-extension). Most published "nice papers" showing effectiveness are probably wrong, or at least, exaggerating the effect. The researcher will not be an exception to the pattern; he will publish if he's "lucky" enough to get colorable results, but that won't actually prove anything other than the fact that journal editors are morons. Actual knowledge requires lots of replication. Might as well plan for it.

    In both cases — the economics of drugs and the economics of medical research and publication — we are badly organized and getting sub-optimal results, and politically paralyzed. So what is new?

  14. "Suppose his study shows the drug does not have the potentially life saving application. Will the army of marketers spread out to prevent its use?"

    In my experience through a person who works with pharma marketing, pharma will push the non-working drug anyway

    The only ethical answer is to not participate in the system and work on the generic.

    In addition to everything else, working with pharm often gives them control of even publications and they typically bury non-positive results and blow even the meagerest positive result way out of proportion in their marketing program.

  15. @James Wimberly: “If doctors are only influenced by Big Pharma marketing and not at all by science, then the US health system is more broken than even I imagined. "

    Indeed. First of all, remind me again what MDs are doing to earn their lifelong-ultra-cushy-income expectations? Forget about the argument that "medical schools and licensing guarantee that only the smartest people can become doctors", if our current doctors aren't expected to use those "smarts" for something as basic as learning about the available treatments from reliable sources.

    Finally: do insurance companies have their own internal NICE offices? Surely Cigna and Kaiser Permanente have the resources, and the financial incentive, to hire "drug reps" (representing generics) just as aggressive as Merck's.

  16. first blog post ever. I am the colleague that proposed the dilemma to keith. Let's clarify the situation with the two drugs. One is an old drug that is off patent, one is a new drug that is on patent that appears to work exactly the same way as the old drug but is absorbed slightly better at high doses. But for our purposes we would expect both drugs to be absorbed the same. let's say we have no a priori reason to think that one would work better than the other. The only difference is that one will make no one any money, and thus will not have a private army of marketers to influence doctors about its possible benefits, while the other, if efficacious, would start down a regulatory road that would include replication of the study, and likely end with a army of marketers pushing the drug at doctors offices. incidentally to sustain lifelong improvement either drug would only need to be taken for three days.

    In any case, if either drug works, I will be doing my best to promote its use to prevent the kind of problems it may be successful in preventing. But one voice in academia is very different than an army of well-funded marketers. The idea that Pharma is now controlling the dialogue, and shaping the dialogue so that we are speaking only about pharmaceutical interventions seems true to me and a generates a separate but related dilemma. we are specifically looking at Pharma to treat this condition because it is the accepted paradigm of those doctors we want to influence to create the better outcome. but we could have looked at a non-Pharma intervention: say hypothetically a backrub. If we discovered that a back rub accomplished the same thing as the drug I think we would wind up with the same dilemma. We would write a paper, it might or might not get ignored, and without the profit incentive to motivate an army of marketers I am not sure it would have any impact. So you could pose the dilemma in the same way: let's say that a priori we have every reason to believe that a back rub and a new on patent medication are equally likely to prevent someone from developing this dreaded disease. You'll need to do a large multiyear study to prove there is a difference if you get the same outcome with either intervention. And you are resource limited. Do you study the back rub and potentially prove that something cheap simple and easy works, or do you study the drug which will cost the system a lot of money but in our current world is more likely to be adopted as widespread practice in part through the work of an army of marketers?

    The academic detailing described above would be a tremendous enhancement to our current medical system, and would likely decrease costs across the board as people were educated about equally and more efficacious alternatives that are off patent and falling by the wayside. I didn't know anyone had ever proposed it in a serious manner. So thanks for educating me

  17. I wouldn't be too hasty to say that the generic and the proprietary product are the same. First of all, there is actually somewhat less regulation of the generics as regards dosage and the like – but secondly and more importantly, the post makes it clear that the proprietary drug is still under intellectual property protection, not merely trademark protection. This means that there must be some difference between it and the generic; people often speak loosely about highly similar drugs when one is off-patent and o e is not. It may be that the two are very similar and affect the same target, they can even have the same active ingredient, but there must be some difference that explains the patent protection. One example I've heard of is where an effective drug is losing its patent protection and its makers patent a new formulation where they've figured out how to make a version that only has to be taken once a day, instead of four times. In such a case, the generic and the proprietary drugs have the same active ingredient, but the once-a-day version is patented, and so more expensive – but then, patient compliance is actually better with the patented once-a-day method.

    All that being said, this is hardly a worst-case scenario. They've got promising data and they managed to get funded for a clinical trial. The scary stories are about the drug companies controlling what gets funded and what gets released, which already isn't going to be the case here. If studying the proprietary version will help more people, then I'd say to do it. If they're having tainted of guilt over the patent holder profiting, they can make extra sure to take no money from the patent holder, and in any case should give them no power over the release of the results. If the generic and the proprietary drugs really are as similar as you imply, then the promotion of the proprietary drug for the condition should also result in patients getting the generic, whether because of some attentive doctors or some penny-pinching insurers.

    And as mentioned in the thread, and without my knowing how well it works in practice, this is an excellent argument in theory for a body like NICE.

  18. Jerry Avorn at Brigham and Women’s in Boston has been a champion of academic detailing for a number of years. He has set up a website for unbiased information about therapeutics at . This site covers a number of high-profile and high-cost interventions, and has links to other online resources. Dr. Avorn has trained pharmacists and nurses to go to doctors’ offices to provide continuing medical education on effective drugs for common conditions; every dollar spent on this effort saves about two dollars in drug expenses.

    The idea is to counter the efforts of sales reps with sound information. I don’t know if they send out attractive pharmacists with low-cut dresses, but that would probably be necessary if they really want to compete with the industry detailers.

    If there are serious deficit hawks on Capitol Hill, they would get behind this effort in a hurry.

  19. Two points:

    Regarding insurers — Kaiser Permanente is probably the largest non-pharmaceutical manufacturer source of privately funded research on effectiveness and cost-effectiveness of drugs in the nation. They generally do not permit manufacturer detailers access to their doctors. This bears true fruit — Kaiser almost always has a lower utilization percentage of new "me too" products, and can bring brand name manufacturers to their knees to command good pricing when they don't think an expensive drug is any better than an older but cheaper branded alternative.

    However, overall, what is spectacularly absent from American medicine is what I call the trusted intermediary who reviews interventions and products on the basis of quality, and neither cost nor profitability. Insurers have data, reams of it, that could be studied to produce these results, but nobody trusts them to act on that data in a way that is neutral to their overall profitability. Manufacturers also have reams of data with the same problem, from the other direction. There is no market solution here because the information needs are too vast, the cost of getting it wrong perceived to be too high, and no one in a position to have BOTH knowledge and trust.

    I do blame doctors for allowing detailers to provide them with information. It is their job to be impartial professionals, not professional whores.

    My working assumption is that most pharmaceuticals, and indeed, a fair amount of medical intervention, is marginally effective if not utterly useless. It's simply obscene that so many people are drinking the medical industrial kool-aid of better living through treatment.

  20. Define "major trial." Also, what Barbara said. Study results published in journal articles may not be completely ignored. I've been told that some major providers (e.g. Kaiser) do pay attention to published research in updating their formularies.

  21. To James Wimberly's question, "How does surgery advance?" Well, in many cases surgery involves surgical implements – stents, joints, etc., and sales people sell those too. E.g., "During many implant surgeries – routine and otherwise – a sales rep from the ortho company is in the operating room with the surgeon to assist with product selection and usage. Usually this help takes the form of simply giving the surgeon the right sized screw when asked. In rare cases the rep might virtually perform the surgery, using a laser pointer over an inexperienced surgeon’s shoulder to position incisions and implant placement. "

  22. NICE is a nice organization — good, smart people. But they do not change clinical practice, they review evidence and produce general guidelines. The US has many, many bodies that do the same thing and do it well. There is nothing NICE does that AHCPR and evidence-based review centers and Cochrane Reviews and QUERI centers don't do also. What changes practice are not NICE's reviews of evidence themselves but changes in the NHS incentive structure that are informed by NICE conclusions — and not incidentally, every day thousands of NHS providers do things that are contrary to what NICE says, and there's isn't anything NICE can do about it.

    In the U.S., the VA sets up financial incentives to encourage evidence-based care (and, gets its drugs more cheaply than Kaiser, BTW) in part because it knows that the information alone doesn't make much impact.

    I used share the faith expressed here that journal articles move the world. I've published a couple hundred, so of course I want to believe that. But it is just that, a belief (Maybe I should not say this, because part of the prestige of a medical school professor rests on this mistaken belief). Producing information alone rarely changes behavior, if it did a million people would quit smoking cigarettes today, everyone would eat 5 helpings of fruit and vegetables a day, no one would be overweight etc.

  23. If I can't study both drugs, I'll study the generic.

    As long as we have to live in the academic environment of Publish or Perish, the results are going to be disseminated in one way or another. NIH not including funding specifically for publication isn't that big a deal, but I wonder why they don't want the results shouted from one rooftop or another.

    As Warren notes, there is a difference between a generic and a drug still under patent protection. Keith acknowledged this in his problem description when he said the proprietary drug is a "copycat" (also called a "Me-too" drug, or "Evergreening", depending on the exact nature of the behavior of the drug company.) The company has its chemists figure out how to make a substitution in a nonfunctional area of the molecule (say, a Cl rather than a H), and hey, presto, New Drug! (For example, flurbiprofen is a derivative of ibuprofen, marketed by Pfizer as Ansaid, Another Non Steroidal AntiInflammatory Drug. Pfizer gets bonus points from me for a devious form of honesty.) Another dodge used is to figure out an economical way to produce the pharmaceutically active isomer of a drug that was previously marketed as a racemic mixture of enantiomers. (The latest example of this that I'm aware of is AstraZeneca's marketing of omeprazole in a racemic mixture of D- and S- isomers as Prilosec, and then producing the S-isomer as Nexium. Cute.)

  24. @ Barbara

    My working assumption is that most pharmaceuticals, and indeed, a fair amount of medical intervention, is marginally effective if not utterly useless. It’s simply obscene that so many people are drinking the medical industrial kool-aid of better living through treatment.

    Um, no. There are definitely cases like this. Some are quite famous, and raise troubling issues of who knew what when. This is particularly true in the case of psychoactives for a whole bunch of reasons. There have been cases where predictions of drug effectiveness that were wrong because those drugs had the expected biological effect but the connection people had made between that biology and that disease turned out to be wrong (HRT, HDL cholesterol, anti-plaque-formation drugs in Alzheimers) – although the latter two of those, like most such cases, were caught in trials. And there have been really quite complicated cases, like the COX2 inhibitors, when genuinely do provide long-term analgesia and escape the known catastrophic effects of long-term usage of drugs targeting multiple COX members (which is why COX2 inhibitors were sought), but also turned out to possibly double the risk of heart attack.

    And, if anything, the stories of unnecessary surgical intervention are worse. Partly this is because it's harder to regulate surgery than drugs, and the ethics of controlled surgical studies are so fraught. In one famous case, people were anesthetized, the backs of their knees were sliced open, and then either they were given an arthroscopy or they were just stitched up. Turned out that arthroscopy offered no benefit, which was worth knowing – but how do you justify anesthetizing and slicing people for no other purpose than to control for the placebo effect?

    But these are all the failure stories. Do you really think that over the last fifty years we aren't Living Better Through (medicinal) Chemistry? Even over the last decade or so, we've seen phenomenal successes in the discovery and optimization of targeted drugs. Look at Gleevec, which would deserve the term "miracle" if that word didn't actually denigrate the intelligent and dedicated work of the researchers who created it. Or look at that anti-b-Raf drug under development that's made the front page of the New York Times several times in the last year because if you've got the right sort of untreatable terminal melanoma it can put you in complete remission for six or nine months.

    Drugs get to be part of "most pharmaceuticals" because they're shown in controlled trials to be effective. Not just effective – more effective than available alternatives. Human bodies and lives and especially minds are difficult, complicated things. Errors are made, and there are certainly some venal people out there. But to claim on the basis of nothing that "most pharmaceuticals [are] marginally effective if not utterly useless" contradicts the facts. It flies in the face of the very notion of medical research.

  25. @Dennis: Perfect description of the me too drug phenomenon, I probably should have used that phrase in the post but I am glad you brought it in here.

    @All: Will post Icy's solution tomorrow.

  26. The Australian researcher Barry Marshall showed in 1985 that many ulcers are caused by a common bug, helicobacter pylori, and can be cured with a course of generic antibiotics. No drug company reps spread this finding, which must have cut billions of any currency from the sales of expensive (but effective) proprietary anti-ulcer drugs like Tagamet and Zantac. So how did the discovery spread? Was it the spectacular exception that proves the rule?

  27. Another random thought. Physicians in the US have to attend a cerain amount of continuing education to preserve their licenses. Via anecdotal reports from family members, it looks as if a lot of this takes the form of Pharma-sponsored junkets in the Caribbean. A battle to recapture continuing medical education so that it's genuine mutual learning based on independent scientific evidence and front-line experience would be worth fighting.

  28. Warren, there is no doubt that I am overly cynical because of my work situation, but truly, many, many drugs provide marginal benefit and many others probably provide even less benefit than that when looked at in view of their potential side effects. I would exempt most blood pressure medications, AIDS medications and a few others from this list. Gleevec is almost the exception that proves the rule in the case of cancer, but Tamoxifen has also provided clear improvements for breast cancer patients. Nonetheless, even for useful drugs, the pattern is for manufacturers to overstate the potential benefits of a drug to the outer limits of credibility. Cancer drugs is one of the few areas where FDA really does stand up to manufacturers on the efficacy of drugs, hence, the action on Avastin. Fosamax and related drugs, along with Epogen, are clear examples of how this trend is not only expensive but actually raises significant risks for many people taking those products. Efforts to treat "preclinical" hypertension or high cholesterol are also an example of this phenomenon. And that doesn't even go to examples like HRT or COX-2 inhibitors.

  29. I am a pharmacologist who has worked in academia, industry, and now back in academia, so I have a fairly good idea of the stimuli that control a lot of the behavior in these two generalized realms.

    One important point that I don’t think has been discussed here yet is the fact that there is a big difference (particularly for physicians) between a study (or several) that show an effect and having an FDA-approved indication for a drug. Like it or not, having an indication is important – many docs (for better or worse) tend not to prescribe medications off-label (i.e., for something for which they are not indicated). Without knowing the specific details of the proposed study (other than “major”), it is hard to tell whether this NIH-funded study could be considered a Phase III trial (typically two are required) and whether the endpoints of this trial would be found acceptable by FDA. Which brings me to my second point…

    Academics are neither incentivized for, nor have the resources for, nor are typically very good at drug development. It is a very complicated and very expensive process during which many things can go wrong. Because there is so much money on the line, companies typically have several meetings with FDA throughout the development of a new drug to make sure that the studies are designed in a way that FDA finds acceptable. I know of very few academic colleagues who have similar discussions with FDA. In an ideal world, there would be real incentives in place for everyone for simply conducting studies and developing therapies that improve public health, but since this conversation is being hosted within the confines of the Reality Based Community, we must acknowledge that this is not the case. Academics engage in research that they think will result in publications and grant money and industry engages in research that they think will result in profit.

    Ultimately, I believe that it is wise to adopt the long view. Patents expire and knowledge takes time to be realized and put into practice. Which would likely happen sooner in this case? If you can get the company to develop the drug and obtain the indication, smart docs will likely use the other drug that is generic anyway and others will prescribe the branded drug that was studied (until it goes off patent and they can prescribe that drug as a generic). How long would that be? If you study the current generic, how likely would that information reach the masses of docs, patients, or general public (perhaps you know important people in the media for example)? How long would that take to happen? How widely would the generic be used (presumably) off-label?

  30. Easy: study the generic. The corporation will still use the study to promote their own product, but for people who can't afford (and aren't willing to be suckered by the corporation) the scientific evidence will point to the cheap, generic drug. It would be immoral to take the alternate path and line the pockets of a corporation when there is an affordable, within-reach product that the majority of people can benefit from.

  31. @James

    Fighting to reclaim CME is a good idea, not least because education isn't marketing.


    True, all of that. But we aren't talking about drug development here, we're talking about a trial on a new indication. And don't let's forget that at least some development goes in government laboratories. Azidothymidine was originally developed under a NCI grant. It wasn't sufficiently active as an anticancer agent, and went on the shelf until Burroughs-Wellcome decided to check it for activity against HIV.

    There isn't any reason that new drugs can't be developed by academics — it happened in the past. Change the incentives and it can happen again. But as you note, the incentive structure for academics has to be changed and the support infrastructure has to be present, too. When I think about it, there isn't any particularly good reason for some ambitious comprehensive health campus (like UCSF, or Oregon's Health Sciences University or Colorado's Health Sciences Center) not to take this on. A block-buster drug patent could support a lot of system-wide activity.

  32. I'm confused to how a company can have a patent on a drug that is already available as a generic? This could only be if the company drug was somehow different from the generic (e.g. a modification, or an enantiomer instead of a racemic mixture).

    I have to agree with Dennis to study both drugs. Any differences or lack of difference in clinical activity between the two should be published.

  33. @ Dennis

    There isn’t any reason that new drugs can’t be developed by academics — it happened in the past.

    Well, sure there's a reason: it's incredibly expensive. The way the system works now, most basic research is funded by the taxpayer, and attempts are made to find ways that findings can inform the search for new treatments. The process of moving from basic research to a candidate, and especially the process of testing a candidate, is incredibly expensive – and rarely successful. Of course, successful drugs are highly lucrative. But they have to pay for all the failures. If you want to propose that research for the discovery of new drugs be nationalized, that's pretty much fine by me – but be aware that in addition to (eventually) stopping the ability of the pharmaceutical companies to gouge, you'll also be imposing huge new expenses on government funding of research. Also, once our government-funded research has discovered a drug, who gets the value of its intellectual property? Is it sold at "market value" (the determination of which for drugs is a mystery to me), with the benefit going to the government? To a combination of the government and the institutions where it was discovered, and maybe the researchers involved? Is it instead sold at a discount to all Americans, with the research costs fully borne by all taxpayers? At what price should it be sold in Europe? This isn't as simple as you seem to think.

    @ DH

    This could only be if the company drug was somehow different from the generic (e.g. a modification, or an enantiomer instead of a racemic mixture).

    Indeed; this was noted above. But it's not obvious whether the term "generic version" was being used loosely (probably do the same thing more or less the same way) or precisely (the exact same amount of the exact same active ingredient), or where on that continuum this case lies. Some of the very small differences the drug companies make in order to have a newly patentable version of an old drug might actually change the drug's properties, even in unexpected ways.

    I have to agree with Dennis to study both drugs.

    In today's funding environment it may be a small miracle they got the funding to do a controlled study of one drug. It's already been stated that they don't have the resources to study two drugs.

  34. Here's a cheaper alternative to academic detailing: require that any physician who wants to be able to prescribe a given class of drugs take and pass, once every three years, an objective-answer exam about that class of drugs.

  35. Yet another point. The frame of reference given for the dilemma is the USA. For medical research, this can't be right. The pharma marketing issue is largely limited to thr US; elsewhere, patients will largely be limited to the generic by price or regulation. American doctors (says Keith) don't get their prescribing information from the NEJM, a fortiori doctors elsewhere. But we can be sure that it's carefully studied by NICE and its counterparts in Brazil and India. Information about the generic therefore has a higher short-term worldwide payoff than about the me-too drug. You would need an implausibly high discount on the welfare of foreigners to overcome this. And by studying the generic, you are nudging the dysfunctional US system in the right direction, rather than reinforcing failure.

  36. Requiring doctors to pass a written test if they want to prescribe a “class of drugs” would create a host of new problems. Who would define the classes of drugs? Are antihypertensives a class? Or should these be subdivided into separate categories of beta blockers, ACE inhibitors, calcium channel antagonists, angiotensin receptor blockers, and diuretics? Would a doctor have to pass the test for each “class” separately? If a doctor wanted to prescribe a tricyclic antidepressant, would he/she have to pass the section on SSRIs? Are antimicrobials a class, or should this be subdivided into antibacterial, antifungal, and antiviral drugs?

    Would these tests be done at the time of board recertification, or would they be something separate?

    And how should differences of opinion and practice pattern be handled? Should cautious practitioners who often want to observe their patients for a while be punished if the correct test answers are defined by examiners who favor earlier and more aggressive intervention? Should the test measure factual knowledge or clinical judgment?

    Academic detailing may create some problems, but it acknowledges and discusses the uncertainties of current best evidence, and has enough flexibility to expand the mindsets of doctors, rather than constrict them, which happens if they have to go and pass one more damn test (often by having to guess which answer the examiner wants to hear). An "objective-answer" exam of drug classes would fail to accomplish many important goals.

  37. @ Warren

    Color me unconvinced about the expense of Phase III trials. I suspect that Pharma is as creative in its accounting as Hollywood is. Of course Phase III trials are expensive, and I'm perfectly willing to have the NIH and NSF putting up more money for pharmaceutical development. As to the IP rights once a drug is developed, tested and released, that's easy. The IP rights are in the public domain if the drug was developed with NSF/NIH money. If Baylor Med develops a drug with Baylor Foundation money, then the IP belongs Baylor. I'm not sure what happens if UCSF develops a drug with a mix of public and Foundation monies.

    One of the problems with our current system is known in the meta-analysis gang as the file-drawer problem. We don't know about the drugs that were screened out in the early phases, because those things don't get published or submitted to the FDA. With public funding, that information is more available. Azidothymidine (AZT) was initially developed as a cancer chemotherapy agent (there was a theory that cancer was caused by retroviruses), and dropped because it showed no promise as a chemotherapy agent. When HIV came along, someone at Burroughs-Wellcome remembered that there were some tests done on nucleoside analogs as reverse transcriptase inihibitors. They pulled AZT off the shelf, and were allowed to patent a drug developed (and run through Phase II trials, I think) with public money.

    @ Ed

    Defining "classes" of drugs is fairly straightforward. Every drug has a physiological target, and you define classes by the target. So, beta blockers would be a class of drugs, serotonin reuptake inhibitors would be another, and so forth. Antimicrobials again have particular targets, and you define the classes by the target mechanism.

    Alternatively, you could define "classes" of drugs very narrowly, as each drug and its mimics. So, ibuprofen would be paired up with all the NSAIDs. If you want to prescribe those, you need to know something about the drug and all (or maybe most) of its mimics, particularly the generic forms.

    I don't think you get too deeply into clinical practice, just measure their factual knowledge about the drugs, their mechanism of action, the common (and dangerous uncommon) side effects and contraindications. Yeah, I think knowing when not to use it is more important than knowing when to use it.

    The wisdom of dropping another certification burden on physicians is questionable, though. I think I'd prefer to outlaw marketing junkets posing as continuing medical education. Put that stuff back into the medical education system, where it belongs, and you have a place for academic detailing to do its job.

  38. Right, Dennis, but if the test is reduced to factual knowledge about drugs, you get to answer questions like the pH in which aminoglycosides are effective; this does not really get to the kinds of questions that are best addressed through discussion rather than by blackening the correct box with a number 2 pencil. Defining drug classes by the physiological target is conceptually sound, but if you split the classes into the various receptors to which they may bind, the result is a theater of the absurd. I think we agree that judgment is the main thing to support, and that is not supported by standardized testing.

  39. @ Ed,

    I think it could be done, but I'm not sure doing it is wise. My preference would be to stop counting marketing junkets paid for by Pfizer and Friends for CME credit. I've seen the invitations to these junkets, and they are invariably held at a really nice resort with plenty of time to do things that aren't medical education. The presenters are generally people with close ties to the sponsoring organization and the costs to the physician don't generally approach the actual cost of the vacation. Stop counting that stuff, and put actual clinical faculty on the dais to discuss issues in current clinical practice and you've got some education going on.

  40. Yeah, well phase III is so expensive that no start up can afford it, which means that they develop the drug and get it through Phase I and maybe Phase II and then sell out to a big Pharma company. Believe it.

    However, in this case there is an interesting bit of ju-jitsu. If the generic shows a benefit, then most certainly the company that owns the proprietary drug will test it by themselves to capture at least part of the market.

  41. Dennis and I are in near perfect agreement about the merits of unbiased academic CME (although big Pharma has made inroads into the universities as well). CME of the kind he describes would probably be more cost-effective than academic detailing in the offices of individual doctors.

    It should not be categorically assumed that journal articles do not influence clinical practice. For example, consider this article, published in JAMA in 1903, suggesting that arsenic may not be good for us after all; most practicing physicians are in agreement:

    Within the last few years much has been published in regard

    to certain special arsenical compounds and their therapeutic

    value. These compounds, cacodylates, were found to be practically

    non-toxic in large doses, and it was claimed that they

    produce many of the therapeutic results of the basal element.

    Some recent investigators, however, have thrown discredit on

    the therapeutic efficacy of these drugs…therefore,

    [they are] incapable of exerting remedial or therapeutic effects

    of the commonly used compounds of this substance. It seems

    now probable that both the cacodylates and this more recent

    substitute will pass into disuse unless some special properties

    not those of arsenic are absolutely determined. The enthusiasm

    with which these were at first received appears to have

    been unwarranted by fuller experience and may be taken as

    another instance of ill-judged medical faith in a remedy which

    is often one of the misfortunes of the profession.

    JAMA. 1903;40:651-652

Comments are closed.