4 ms·

> Or is there a case where one could "pass" this test, but still "lose" against a placebo? The main issue is that a lot of the existing drugs aren't better tha

> Or is there a case where one could "pass" this test, but still "lose" against a placebo? The main issue is that a lot of the existing drugs aren't better than placebos: - Some were never required to be tested for efficacy by the FDA before approval. - Some were tested for whatever reason., but the methodology used by testing was flawed for whatever reason. - In many the data was blatantly faked or manipulated by the company running the test. In general NIH tests show much lower efficacy for the same drugs as in pharma trials sponsored by the drug manufacturer. There is tons of research on this. However, there are other tricks that pharma companies use when testing against existing drugs. One of the main ones is using non-effective doses of the older drugs... E.g. by using a higher than normal dose of the older drug in order to make it look like the newer drug has less side effects. This is how the newer generation of anti-psychotics got approved, and it's also how Nexium got approved even though it's basically chemically identical to Prilosec.

refurb 12y ago
Some were never required to be tested for efficacy by the FDA before approval Yes they were used before the FDA had a formal approval process, but your comments doesn't make sense. If the current standard of care is a drug and you test against it, aren't you showing it's more effective?
- Alex3917 12y ago
  > If the current standard of care is a drug and you test against it, aren't you showing it's more effective? The main issue is that a drug can be more effective than an older drug but still be less effective than taking no drug at all. But given all the problems, both theoretical and practical, with current drug trials, there is pretty much no guarantee of anything. Just because something makes sense on paper doesn't mean that it's actually true. There are a number of books about the issues with drug trials... Bad Pharma by Ben Goldacre, The Truth About Drug Companies by Marcia Angell, Overdosed America, etc. All of these books cover largely different issues, so imho it's worth reading them all.
  - refurb 12y ago
    The Truth About Drug Companies by Marcia Angell No offense, but Marcia Angell has a tenuous grasp in terms of knowledge of the pharmaceutical industry. I would take her opinions with a grain of salt.[1] [1]http://www.forbes.com/sites/johnlamattina/2012/12/14/marcia-angells-attacks-on-pharma-have-lost-all-credibility/ http://www.forbes.com/sites/johnlamattina/2012/12/14/marcia-...
    - Alex3917 12y ago
      If you had read the actual book, you'd know that this article is completely mischaracterizing her arguments. From the article: > Angell’s attack came at the same time that the FDA was reporting that much like 2011, 2012 has proven to a great year for patients with the approval of 35 innovative drugs. Here is the actual argument from the book: "In the five years from 1998 through 2002 [...] 415 new drugs were approved -- an average of 83 per year. Of those, 133 (32 percent) were new molecular entities. The others were variations of old drugs. And of those 133, only 58 were priority review drugs. That averages out to no more than 12 innovative drugs per year, or 14 percent of the total." "Over the four years beginning in 2000, there were just 32 innovative drugs out of a total of 314 approved drugs. Even in this small collection, only seven came from the top ten U.S. drug companies." The book was published ten years ago, but if you look at the most recent data from the FDA then it's similar: In the five years between 2007 and 2011, a total of 119 New Molecular Entities were approved by the FDA. Of these, only 50 were granted priority review, meaning they were classified by the FDA as having the potential to provide a therapy where none exists or providing a therapy which may be potentially superior to existing therapy. Of these 50 priority NMEs, 30 were orphan drugs, meaning they were approved for diseases that affect 200,000 or fewer Americans. Of these, it's not clear how many (if any) were created by large American pharma companies. [1] http://www.fda.gov/Drugs/DevelopmentApprovalProcess/HowDrugsareDevelopedandApproved/DrugandBiologicApprovalReports/ucm121136.htm http://www.fda.gov/Drugs/DevelopmentApprovalProcess/HowDrugs...
      - refurb 12y ago
        Exactly! This quote is entirely wrong: ""In the five years from 1998 through 2002 [...] 415 new drugs were approved -- an average of 83 per year. Of those, 133 (32 percent) were new molecular entities. The others were variations of old drugs. And of those 133, only 58 were priority review drugs. That averages out to no more than 12 innovative drugs per year, or 14 percent of the total."" Just because it didn't get priority review, doesn't mean it's not innovative. It just shows how little Angell understands the industry.
taeric 12y ago
Ok, so it is less about them not specifically testing against a placebo, and more that they just do shoddy tests in the first place. I think that makes sense.
pluma 12y ago
Worse yet, I recall a study that found placebos are actually getting better over time. As in: drugs that have been tested successfully before fail in new tests because the placebos perform better than before. I'm not entirely sure what the plausible explanation for that one would be. Other than that the testing methods have changed to create a bias in favour of the placebos being successful.
- Alex3917 12y ago
  > Worse yet, I recall a study that found placebos are actually getting better over time. I think that was just a Jonah Lehrer thing. (I.e. his article The Truth Wears Off in The New Yorker.) What's actually happening is more likely that if a new class of drugs makes it onto the market and becomes popular enough, then it eventually gets studied by people who don't have an economic incentive in 'proving' that the drug is actually effective. Once you stop cherry picking the subjects of the trials, covering up bad data, and using dubious statistical methods, the drugs start looking much less effective than they did originally.

refurb 12y ago

Some were never required to be tested for efficacy by the FDA before approval Yes they were used before the FDA had a formal approval process, but your comments doesn't make sense. If the current standard of care is a drug and you test against it, aren't you showing it's more effective?

Alex3917 12y ago
> If the current standard of care is a drug and you test against it, aren't you showing it's more effective? The main issue is that a drug can be more effective than an older drug but still be less effective than taking no drug at all. But given all the problems, both theoretical and practical, with current drug trials, there is pretty much no guarantee of anything. Just because something makes sense on paper doesn't mean that it's actually true. There are a number of books about the issues with drug trials... Bad Pharma by Ben Goldacre, The Truth About Drug Companies by Marcia Angell, Overdosed America, etc. All of these books cover largely different issues, so imho it's worth reading them all.
- refurb 12y ago
  The Truth About Drug Companies by Marcia Angell No offense, but Marcia Angell has a tenuous grasp in terms of knowledge of the pharmaceutical industry. I would take her opinions with a grain of salt.[1] [1]http://www.forbes.com/sites/johnlamattina/2012/12/14/marcia-angells-attacks-on-pharma-have-lost-all-credibility/ http://www.forbes.com/sites/johnlamattina/2012/12/14/marcia-...
  - Alex3917 12y ago
    If you had read the actual book, you'd know that this article is completely mischaracterizing her arguments. From the article: > Angell’s attack came at the same time that the FDA was reporting that much like 2011, 2012 has proven to a great year for patients with the approval of 35 innovative drugs. Here is the actual argument from the book: "In the five years from 1998 through 2002 [...] 415 new drugs were approved -- an average of 83 per year. Of those, 133 (32 percent) were new molecular entities. The others were variations of old drugs. And of those 133, only 58 were priority review drugs. That averages out to no more than 12 innovative drugs per year, or 14 percent of the total." "Over the four years beginning in 2000, there were just 32 innovative drugs out of a total of 314 approved drugs. Even in this small collection, only seven came from the top ten U.S. drug companies." The book was published ten years ago, but if you look at the most recent data from the FDA then it's similar: In the five years between 2007 and 2011, a total of 119 New Molecular Entities were approved by the FDA. Of these, only 50 were granted priority review, meaning they were classified by the FDA as having the potential to provide a therapy where none exists or providing a therapy which may be potentially superior to existing therapy. Of these 50 priority NMEs, 30 were orphan drugs, meaning they were approved for diseases that affect 200,000 or fewer Americans. Of these, it's not clear how many (if any) were created by large American pharma companies. [1] http://www.fda.gov/Drugs/DevelopmentApprovalProcess/HowDrugsareDevelopedandApproved/DrugandBiologicApprovalReports/ucm121136.htm http://www.fda.gov/Drugs/DevelopmentApprovalProcess/HowDrugs...
    - refurb 12y ago
      Exactly! This quote is entirely wrong: ""In the five years from 1998 through 2002 [...] 415 new drugs were approved -- an average of 83 per year. Of those, 133 (32 percent) were new molecular entities. The others were variations of old drugs. And of those 133, only 58 were priority review drugs. That averages out to no more than 12 innovative drugs per year, or 14 percent of the total."" Just because it didn't get priority review, doesn't mean it's not innovative. It just shows how little Angell understands the industry.

taeric 12y ago

Ok, so it is less about them not specifically testing against a placebo, and more that they just do shoddy tests in the first place. I think that makes sense.

pluma 12y ago

Worse yet, I recall a study that found placebos are actually getting better over time. As in: drugs that have been tested successfully before fail in new tests because the placebos perform better than before. I'm not entirely sure what the plausible explanation for that one would be. Other than that the testing methods have changed to create a bias in favour of the placebos being successful.

Alex3917 12y ago
> Worse yet, I recall a study that found placebos are actually getting better over time. I think that was just a Jonah Lehrer thing. (I.e. his article The Truth Wears Off in The New Yorker.) What's actually happening is more likely that if a new class of drugs makes it onto the market and becomes popular enough, then it eventually gets studied by people who don't have an economic incentive in 'proving' that the drug is actually effective. Once you stop cherry picking the subjects of the trials, covering up bad data, and using dubious statistical methods, the drugs start looking much less effective than they did originally.