Page 583 of 725 FirstFirst ... 83483533573581582583584585593633683 ... LastLast
Results 5,821 to 5,830 of 7244

Thread: Coronavirus thread

  1. #5821
    Site Supporter
    Join Date
    Jul 2017
    Location
    Texas
    Quote Originally Posted by whomever View Post
    As a senior stats major, we ran clinics to help MS and PhD students from other disciplines analyze their data. It was really sad to have to explain to someone that the data from their three years of fieldwork wasn't useful for proving or disproving their hypothesis.
    Did you then tell them to find another hypothesis that would fit their data? I would see school officials who should have know better use gain scores within groups to claim achievement progress. Kids who already read at a high level are unlikely to score much higher, and kids with lower scores can readily show bigger increases. Raw gain score data require a different treatment than making simple comparisons.

  2. #5822
    Site Supporter 0ddl0t's Avatar
    Join Date
    Feb 2019
    Location
    Jefferson
    The classic example of doctors & statistics:


    Schonberger and Grayboys(1978) presented 20 house officers, 20 fourth year medical students and 20 attending physicians from Harvard Medical School with the following question:

    "If a test to detect a disease whose prevalence is 1/1000 has a false positive rate of 5%, what is the chance that a person found to have a positive result actually has the disease, assuming you know nothing about the person's symptoms or signs?"

    The most common response given by almost half of the participants was 95%. The average answer was 56% and only 11 participants gave the appropriate response of 2%.

    Covid example I see misunderstood almost daily:

    Of the ~2 million Californians being tested every 14 days, ~6% are coming back positive which we assume is accurate. Abbot has a fast test that has 1% false positives, but 10% false negatives. If we use Abbot's fast test instead, how accurate will our positive count be for the total sample?

    So of that (non random) 2,000,000 sample, we expect 120,000 to actually be infected. The FAST test would yield 20,000 false positives and 12,000 false negatives, so it would report a total of 128,000 infections - 7% too many!

    Yet I've repeatedly seen public health officials claim the test's high false positive rate would skew their total infection count DOWN...

  3. #5823
    Quote Originally Posted by willie View Post
    Did you then tell them to find another hypothesis that would fit their data?
    Nope. That's called 'p-hacking', and it's kind of a no-no, although commonly done.

    E.g. https://www.npr.org/sections/thesalt...ns-for-science

    "The gold standard of scientific studies is to make a single hypothesis, gather data to test it, and analyze the results to see if it holds up. By Wansink's own admission in the blog post, that's not what happened in his lab.

    Instead, when his first hypothesis didn't bear out, Wansink wrote that he used the same data to test other hypotheses. "He just kept analyzing those datasets over and over and over again, and he instructed others to do so as well, until he found something," "

  4. #5824
    Site Supporter
    Join Date
    Nov 2012
    Location
    Erie County, NY
    That's a giant debate. When funding and tenure depends on publication, you can't publish something that found no effect according to some p level. I was on a master's committee of student who ran the experiment she and the prof designed. They didn't find an effect. The study was well done and enough, for a master's thesis that would never see the light of day in publication. However, the chair of her committee was desperate for a publication and kept making her analyze and analyze to find something. It was going to screw up getting her degree and job. Finally, myself and her department chair took said prof aside and said, You are not going to promoted if you keep this up. Approve the thesis. We will go for ethics charges if need be.

    Student graduated.

    Anyway, tenure and academic jobs are destroyed by the virus. There won't be new hiring for years to come. Adjuncts will be reduced (those poor folks desperately trying to hang on).

    College athletics are toast for a bit.

  5. #5825
    Site Supporter
    Join Date
    Mar 2012
    Location
    Sierra Nevada Mtns, CA
    Quote Originally Posted by whomever View Post
    As a senior stats major, we ran clinics to help MS and PhD students from other disciplines analyze their data. It was really sad to have to explain to someone that the data from their three years of fieldwork wasn't useful for proving or disproving their hypothesis.
    We need to hang out, I got lots of data and no conclusions.

  6. #5826
    Site Supporter
    Join Date
    Jun 2020
    Location
    Missouri
    Quote Originally Posted by Glenn E. Meyer View Post
    That's a giant debate. When funding and tenure depends on publication, you can't publish something that found no effect according to some p level. I was on a master's committee of student who ran the experiment she and the prof designed. They didn't find an effect. The study was well done and enough, for a master's thesis that would never see the light of day in publication. However, the chair of her committee was desperate for a publication and kept making her analyze and analyze to find something. It was going to screw up getting her degree and job. Finally, myself and her department chair took said prof aside and said, You are not going to promoted if you keep this up. Approve the thesis. We will go for ethics charges if need be.

    Student graduated.

    Anyway, tenure and academic jobs are destroyed by the virus. There won't be new hiring for years to come. Adjuncts will be reduced (those poor folks desperately trying to hang on).

    College athletics are toast for a bit.
    You're doing the Lord's work, man. Somebody needs to stand up for grad students against crummy professors.

  7. #5827
    THE THIRST MUTILATOR Nephrology's Avatar
    Join Date
    Sep 2011
    Location
    West
    An article by Matt Taibbi on the pharmaceutical industry & COVID-19

    Big Pharma's COVID-19 Profiteers

    Americans reacted in horror five years ago when a self-satisfied shark of an executive named Martin Shkreli, a.k.a. the “Pharma Bro,” helped his company, Turing Pharmaceuticals, raise the price of lifesaving toxoplasmosis drug Daraprim from $13.50 to $750 per pill. Shkreli, who smirked throughout congressional testimony and tweeted that lawmakers were “imbeciles,” was held up as a uniquely smug exemplar of corporate evil. On some level, though, he was right to roll his eyes at all the public outrage. Although he was convicted on unrelated corruption charges, little about his specific attitudes toward drug pricing was unusual. Really, the whole industry is one big Shkreli, and Covid-19 — a highly contagious virus with unique properties that may require generations of vaccinations and booster shots — looms now as the ultimate cash cow for lesser-known Pharma Bros.
    I thought this piece was a little weaker than some of his others, as I sort of feel like his take on a few smaller details is a little bit off (i.e. the FDA clinical trials process vs. rest of world; I would argue we actually have more stringent approvals process than the EU for example) but his general message isn't wrong.

    The United States gov't spends far more than any other nation on biomedical research (in fact, your federal taxes have paid my salary and subsidized my training every year since 2013), and it is true that the expectation is that the basic scientific discoveries that are funded by the US gov't will be turned into a commercial product by the pharmaceutical industry. I don't think this is a wholly terrible strategy - in theory it encourages the kind of risk-taking necessary in basic science investigation that would be unpalatable do a private business - but it absolutely means that we are paying twice for drugs that are sold more cheaply to other countries.

  8. #5828
    Site Supporter
    Join Date
    Jun 2020
    Location
    Missouri
    Quote Originally Posted by Nephrology View Post
    An article by Matt Taibbi on the pharmaceutical industry & COVID-19

    Big Pharma's COVID-19 Profiteers



    I thought this piece was a little weaker than some of his others, as I sort of feel like his take on a few smaller details is a little bit off (i.e. the FDA clinical trials process vs. rest of world; I would argue we actually have more stringent approvals process than the EU for example) but his general message isn't wrong.

    The United States gov't spends far more than any other nation on biomedical research (in fact, your federal taxes have paid my salary and subsidized my training every year since 2013), and it is true that the expectation is that the basic scientific discoveries that are funded by the US gov't will be turned into a commercial product by the pharmaceutical industry. I don't think this is a wholly terrible strategy - in theory it encourages the kind of risk-taking necessary in basic science investigation that would be unpalatable do a private business - but it absolutely means that we are paying twice for drugs that are sold more cheaply to other countries.
    You probably know more than me, but my understanding is that we (in the US) can't pay the cheap prices other nations do because our expensive prices foot the R&D bill. If we didn't pay high prices, a lot of stuff wouldn't get made.

    I don't know where I picked up this viewpoint, though. Take that for what it's worth.

  9. #5829
    THE THIRST MUTILATOR Nephrology's Avatar
    Join Date
    Sep 2011
    Location
    West
    Quote Originally Posted by Bio View Post
    You probably know more than me, but my understanding is that we (in the US) can't pay the cheap prices other nations do because our expensive prices foot the R&D bill. If we didn't pay high prices, a lot of stuff wouldn't get made.

    I don't know where I picked up this viewpoint, though. Take that for what it's worth.
    Really the reason we pay more than the rest of the world is because the pharmaceutical industry knows that the US consumer has less leverage than the average European/Canadian/etc consumer. This is because in France, for example, drug prices are negotiated by the French gov't, and you have to negotiate with them if you want to sell your product in France (or the EU, or whatever).

    Because healthcare systems in the USA are fragmented (reducing their negotiation leverage) and because insurance is not a right afforded to Americans, they are able to charge us more for the same product, as they have the upper hand in negotiating a final price for the sales of their drugs. There is also I'm sure a lot of complexity in the way that bulk drug deals are written to get them to your local CVS, but I don't know anything about that.

    The "front-end" development of major pharmaceutical drugs like Keytruda or Gleevec happens at research universities and academic medical centers, the vast majority of the time. By "front-end," I am referring to basic scientific discoveries that use cells, animals, test tubes, etc etc to uncover the basic biological mechanisms of human disease. The vast majority of this work is funded by the NIH (and to a lesser extent, private charities) that award grants to scientists who have professorships at these universities.

    Once that work has "ripened," if you will, to the point that the pharmaceutical industry believes it might be an idea that would be profitable to develop as a novel drug, then industry steps in. They do a lot of the work that is involved in polishing the idea up and making it ready for production by a GMP facility so the company can apply for an NDA and begin the FDA clinical trials process.

    There is a substantial amount of money and effort involved in just getting an idea from scientific discovery to ready for Phase 1 trials, so I don't mean to imply that the industry is getting a free ride here. There is a substantial (arguably increasing) risk that an attractive basic scientific concept will not pan out in human beings, for lots of reasons I can elaborate another time. This would be R&D end you're referring to in your question. How that is structured into their pricing, I do not really know.

    However, the fact of the matter is, all of the risk and up-front effort for a majority of drugs approved in the 21st century was done at in an academic research lab, most likely on the American taxpayer's dime. That we pay more for the same drug than citizens of other countries, despite the fact that we likely subsidized most of the basic scientific research upfront, is a curious consequence of the way our healthcare system is organized.

    I'm very proud of both our basic scientific research workforce as well as the world-class talent available in our healthcare system, for what it's worth, as someone with feet firmly in both pools; but I still see a lot of room for improvement.

    Quote Originally Posted by whomever View Post
    Nope. That's called 'p-hacking', and it's kind of a no-no, although commonly done.
    EXTREMELY commonly done.
    Last edited by Nephrology; 08-13-2020 at 03:17 PM.

  10. #5830
    I sometimes think we should say that drug companies can sell their product for whatever they like (so they can cover R&D) but must sell it for the same price to all comers (with maybe some things like volume discounts). There is no reason that, say, the Canadian gov't ought to be paying less than Medicare does.

    If we want to sell at the marginal cost of production as charity to very poor countries, OK, call it foreign aid. But all first world countries ought to be paying the same.

User Tag List

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •