Skip to main content

View Diary: Health Care Friday (140 comments)

Comment Preferences

  •  Proof: (0+ / 0-)

    A statistical study shows there is "no significant benefit" for women to have a mammogram before age 50.

    However, out of the set of all women, some women die of breast cancer before age 50. It is known that early detection of breast cancer will save lives. Therefor, yearly tests for woman under age 50 will reduce deaths from breast cancer.

    So we seemingly have a contradiction, but not really. The conclusion drawn from deductive logic is irrefutable. The problem is we cannot have an honest discussion of the economics of performing fewer tests in terms of the lives that can be saved. Life is supposedly sacred and priceless, but when it comes to health care we all know that is a lie.

    As standard practice in medical studies and journal articles, I have found multiple problems with the blind black box application of the hypothesis test. Rarely are all confounding variables identified and accounted for. The IID (independent and identically distributed) assumption so commonly used is almost always violated by significant temporal and spatial correlation. Well-known non-parametric analysis is rarely employed, so that the distribution of the underlying population is assumed as something but not realistic. Finally, the results of hypothesis tests are easily manipulated to give a desired result, or an accidental but incorrect result. This is known as the Hypothesis Test Fallacy. Look it up.

    How did the stones vote this time? They voted for hardness and few words. And the men? They voted against themselves and for fire which they could not control.

    by lightfoot on Fri Nov 20, 2009 at 08:25:45 PM PST

    [ Parent ]

    •  if you come in with an agenda like you did (0+ / 0-)

      and a pre-conceived notion (which, by the way, is a false one in this case), you'll never learn a blessed thing in your life. It's isn't the statistics that are at fault, it's your own bias, I'm afraid.

      Just saying.

      There are much better, more realistic and more fair analyses than yours. Try:

      New Mammogram Guidelines Are Causing Confusion, But Here’s Why They Make Sense

      Really rethinking breast cancer screening

      Evidence Based Medicine and Reform

      just for starters.

      Any study can be challenged in terms of the stats and the universe of patients studied. But assumptions about

      The problem is we cannot have an honest discussion of the economics of performing fewer tests in terms of the lives that can be saved. Life is supposedly sacred and priceless, but when it comes to health care we all know that is a lie.

      have nothing to do with facts, other than you are the one who can't be honest about what the data shows and need to inject some other framework to justify outrage that doesn't belong in the discussion.

      If you want bumper stickers and death panels, try Sarah Palin's campaign and good luck to you. In the meantime, the question on the table is what works best for women of that age. The guidelines say it's not clear, and that women between 40 and 50 should have a discussion about it with their own doc. That's what the guidelines say. Go read them.

      "Politics is the art of looking for trouble, finding it everywhere, diagnosing it incorrectly and applying the wrong remedies." - Groucho Marx

      by Greg Dworkin on Sat Nov 21, 2009 at 04:50:35 AM PST

      [ Parent ]

      •  I will study your links and get back if (0+ / 0-)

        I find something I do not already know.

        Even though you completely failed to refute my logical reasoning, either by showing the premises to be false or by showing a logical flaw in the conclusion (pretty much an undergraduate exercise):

        Out of the set of all women, some women die of breast cancer before age 50. It is known that early detection of breast cancer will save lives. Therefor, yearly tests for woman under age 50 will reduce deaths from breast cancer.

        Your desperate ad hominem attack inviting me to join Palin only belies the weakness of your arguments.

        How did the stones vote this time? They voted for hardness and few words. And the men? They voted against themselves and for fire which they could not control.

        by lightfoot on Sat Nov 21, 2009 at 05:22:48 AM PST

        [ Parent ]

        •  I don't waste time with bullshit (0+ / 0-)

          if you want to pat yourself on the back for what you rhink is a "win" , go write a diary.

          "Politics is the art of looking for trouble, finding it everywhere, diagnosing it incorrectly and applying the wrong remedies." - Groucho Marx

          by Greg Dworkin on Sat Nov 21, 2009 at 06:00:50 AM PST

          [ Parent ]

          •  Refutation by assertion. Nice. (0+ / 0-)

            The last refuge of scoundrels.

            How did the stones vote this time? They voted for hardness and few words. And the men? They voted against themselves and for fire which they could not control.

            by lightfoot on Sat Nov 21, 2009 at 06:03:12 AM PST

            [ Parent ]

            •  sure thing (0+ / 0-)

              meanwhile your sober "they are deliberately killing women" is a serious discussion? not hardly.

              "Politics is the art of looking for trouble, finding it everywhere, diagnosing it incorrectly and applying the wrong remedies." - Groucho Marx

              by Greg Dworkin on Sat Nov 21, 2009 at 06:05:19 AM PST

              [ Parent ]

              •  Have you heard of the newest trendy phrase? (0+ / 0-)

                "Murder by spreadsheet"? I am sure that its usage is just a passing trend, used by ignorant people who are aligned with Palin.

                No, wait...

                How did the stones vote this time? They voted for hardness and few words. And the men? They voted against themselves and for fire which they could not control.

                by lightfoot on Sat Nov 21, 2009 at 06:09:41 AM PST

                [ Parent ]

                •  "Murder by spreadsheet" sounds like nyceve (0+ / 0-)

                  who I know and have worked with (I was on her panel at YK in Chicago) and has nothing whatever to do with these guidelines, which are evidence based, not insurance based. Just another error you've made.

                  "Politics is the art of looking for trouble, finding it everywhere, diagnosing it incorrectly and applying the wrong remedies." - Groucho Marx

                  by Greg Dworkin on Sat Nov 21, 2009 at 06:15:16 AM PST

                  [ Parent ]

                  •  Evidence-based? (0+ / 0-)

                    Sounds like another step towards science.

                    I pretty much quit commenting in nyceve's diaries a long time ago, although I still read them and rec them. If you go back far enough in the comments you will see I had a fundamental disagreement with her. At least that ended amicably, due to her grace. We agreed to disagree.

                    Have you ever written programs for spreadsheets, or even used them with a full understanding of the underlying mathematics? Just curious.

                    How did the stones vote this time? They voted for hardness and few words. And the men? They voted against themselves and for fire which they could not control.

                    by lightfoot on Sat Nov 21, 2009 at 06:32:57 AM PST

                    [ Parent ]

                    •  the answer is yes (0+ / 0-)

                      I'm an MIT graduate and I am very familiar with the terms and the process, though I am a clinician and not a stats expert and I don't do programming these days.

                      BTW, I've had my very amicable disagreements with nyceve as well ;-) but why did you pick that term (murder by spreadsheet) to inject?

                      "Politics is the art of looking for trouble, finding it everywhere, diagnosing it incorrectly and applying the wrong remedies." - Groucho Marx

                      by Greg Dworkin on Sat Nov 21, 2009 at 06:42:17 AM PST

                      [ Parent ]

                      •  Insurance companies use them to compute (1+ / 0-)
                        Recommended by:
                        DemFromCT

                        statistics. More probably they use "S" or something similar. You are probably right, I picked up the phrase from nyceve.

                        I am sorry that you consider my posts as making me a "jerk". I have stated a logical argument. I only ask for you to consider that much of the field of elementary statistics are based on the fundamental theorem known as Law of Large Numbers. Please consider the implications of this when it is applied to individuals.

                        How did the stones vote this time? They voted for hardness and few words. And the men? They voted against themselves and for fire which they could not control.

                        by lightfoot on Sat Nov 21, 2009 at 06:53:03 AM PST

                        [ Parent ]

                        •  that's an important contribution and thank you (1+ / 0-)
                          Recommended by:
                          lightfoot

                          please understand in return:

                          1. reasonable people can disagree about what the current (insufficient) data says and means
                          1. this is science driven, not insurance driven
                          1. because someone draws a conclusion from the data that you may disagree with or find wanting doesn't mean they are "trying to kill women" or that they don't care about them. I can't begin to tell you how offensive that is.

                          "Politics is the art of looking for trouble, finding it everywhere, diagnosing it incorrectly and applying the wrong remedies." - Groucho Marx

                          by Greg Dworkin on Sat Nov 21, 2009 at 06:58:04 AM PST

                          [ Parent ]

                          •  Granted. (1+ / 0-)
                            Recommended by:
                            DemFromCT

                            It is my opinion that in the past the medical needs of women have been subsumed by the medical needs of men to some degree. I believe it is the case that this approach did not completely change until the 1990's, so this change occurred rather later in my life. If I overreact on the side of the interests of women, then it is due to concern for my wife and younger sister, and in the memory of my sister who passed on.

                            Based on your point 2 I am sorry if I came across too harshly.

                            How did the stones vote this time? They voted for hardness and few words. And the men? They voted against themselves and for fire which they could not control.

                            by lightfoot on Sat Nov 21, 2009 at 07:16:27 AM PST

                            [ Parent ]

                          •  granted for the past, without argument (0+ / 0-)

                            but this isn't the past. the data need to be strong enough to stand on its own. When we don't know we need to say we don't know. That is exactly what the guidelines say.

                            it's disruptive, but it would be dishonest to pretend we KNOW that screening is proven when the opposite is true.

                            "Politics is the art of looking for trouble, finding it everywhere, diagnosing it incorrectly and applying the wrong remedies." - Groucho Marx

                            by Greg Dworkin on Sat Nov 21, 2009 at 07:39:03 AM PST

                            [ Parent ]

                          •  There is a old saying among mathematicians, (1+ / 0-)
                            Recommended by:
                            DemFromCT

                            and that is "nothing can be proven with statistics". Statistics are only used when the data is incomplete, and the nature of the underlying population is unknowable. We use statistics only to make estimations to fill in our ignorance.  Other applications of statistics are abusive.

                            Statistics are always overridden by formal logic and/or set theory wherever possible. In those fields we can design irrefutable proofs of our conclusions, whereas a statistical hypothesis test can yield only probabilities within some confidence interval.

                            This brings me back to my original point. I do not mean to be insulting to you personally, but this is a flaw in the nature of a great deal of medical research. It is why the lay public often hears so much conflicting information, and it is not scientific.

                            I always feel bad when I leave a discussion with someone feeling insulted, it is my soft spot, but ultimately I must remain true to myself and what I know.

                            It is clear to me we cannot agree on this topic, so have a final say if you wish but I am going to let it be.

                            Peace, DemFromCT.

                            How did the stones vote this time? They voted for hardness and few words. And the men? They voted against themselves and for fire which they could not control.

                            by lightfoot on Sat Nov 21, 2009 at 10:30:01 AM PST

                            [ Parent ]

                          •  good health to your family (0+ / 0-)

                            and understand that when docs get it wrong ("it" is sometimes the diagnosis, sometimes the math) it's usually from trying to do the right thing.

                            Peace, lightfoot.

                            "Politics is the art of looking for trouble, finding it everywhere, diagnosing it incorrectly and applying the wrong remedies." - Groucho Marx

                            by Greg Dworkin on Sat Nov 21, 2009 at 11:36:42 AM PST

                            [ Parent ]

            •  and people who disagree with you are scoundrels? (0+ / 0-)

              a complete waste of time. if you want to continue documenting why, go right ahead.

              You started with a chip on your shoulder, and you're ending with a chip on your shoulder.  You documented it well. There's nothing objective or logical in your discussion. If you want to go on about citing "dead woman" with condescending remarks about how only you understand the literature and statistics, you can talk to yourself to your heart's content.

              Or, go here and comment. Maybe he's got more patience for your bullshit than I do.

              "Politics is the art of looking for trouble, finding it everywhere, diagnosing it incorrectly and applying the wrong remedies." - Groucho Marx

              by Greg Dworkin on Sat Nov 21, 2009 at 06:10:37 AM PST

              [ Parent ]

              •  Refutation by Assertion is a Logical Fallacy. (0+ / 0-)

                If that is the best you can do, maybe it is you who carries the sequoia.

                It is weighting you down.

                How did the stones vote this time? They voted for hardness and few words. And the men? They voted against themselves and for fire which they could not control.

                by lightfoot on Sat Nov 21, 2009 at 06:15:05 AM PST

                [ Parent ]

                •  bye (1+ / 0-)
                  Recommended by:
                  lightfoot

                  next time you come charging in acting like a jerk, don't be suprised if you get treated like one.

                  If you don't like it, write a diary. That's what they are for.

                  "Politics is the art of looking for trouble, finding it everywhere, diagnosing it incorrectly and applying the wrong remedies." - Groucho Marx

                  by Greg Dworkin on Sat Nov 21, 2009 at 06:23:38 AM PST

                  [ Parent ]

      •  From your second link: (0+ / 0-)

        The first thing that women need to understand is that these recommendations are for asymptomatic women at average risk for breast cancer undergoing routine screening for breast cancer.

        So the claim here is that women under 50 do not need annual screening for breast cancer if they are of "average risk"?

        Next sentence:

        They are not for women judged to be at high risk due to genetic mutations, strong family history, or other factors producing a high risk for breast cancer. Neither are they for women who are not completely asymptomatic.

        Ohh, so women with "high risk" should have annual testing?

        Unless "average risk" is zero, then these are just weasel words for "more women under 50 will die of breast cancer without annual testing".

        It seems that in "Evidence-based Medicine" dead women are considered to be insignificant statistics. Unless of course it is our "significant other", or our mother, our sister or friend.

        Ignorance in medical sciences promotes the confusion between the statistical definition of "significant" vs. the common meaning of the word.

        How did the stones vote this time? They voted for hardness and few words. And the men? They voted against themselves and for fire which they could not control.

        by lightfoot on Sat Nov 21, 2009 at 05:45:11 AM PST

        [ Parent ]

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site