In which I critique an article that critiques articles

It be all meta up in here.

So my mom sent me this: Lies, Damned Lies, and Medical Science. Go on and read the article.  I’ll wait.

I read it and had some thoughts.

What is in there is not inconsistent with believing in the results of good science.  The problem is evaluating what is good science, and not getting it oversimplified or distorted.  This part:

“In poring over medical journals, he was struck by how many findings of all types were refuted by later findings.  […]  And they sometimes make headlines, as when in recent years large studies or growing consensuses of researchers concluded that mammograms, colonoscopies, and PSA tests are far less useful cancer-detection tools than we had been told; or when widely prescribed antidepressants such as Prozac, Zoloft, and Paxil were revealed to be no more effective than a placebo for most cases of depression; or when we learned that staying out of the sun entirely can actually increase cancer risks; or when we were told that the advice to drink lots of water during intense exercise was potentially fatal; or when, last April, we were informed that taking fish oil, exercising, and doing puzzles doesn’t really help fend off Alzheimer’s disease, as long claimed. Peer-reviewed studies have come to opposite conclusions on whether using cell phones can cause brain cancer, whether sleeping more than eight hours a night is healthful or dangerous, whether taking aspirin every day is more likely to save your life or cut it short, and whether routine angioplasty works better than pills to unclog heart arteries.”

is, to my mind, probably a symptom of media oversimplifying complex studies and of consumers and doctors not understanding the primary sources.  For example, the part about water and exercise?  It’s STILL a good idea to drink water while exercising, but not to the point of depleting your electrolytes and imbalancing yourself.  Like, duh.  Nothing in excess.  Athletes have known this for a long time.

The antidepressant pills have *always* been known to be not that great (though this wasn’t always publicized well), and some of them may even increase suicidal thoughts.  But for some people, sometimes, they do work — even if it is a placebo effect, that still helps people.  There is *still* no good explanation for how their biochemical changes ameliorate depression, but sometimes it works.  Sometimes not.  This isn’t new.

The mammograms-aren’t-a-panacea thing I’ve been hearing for several years, largely from feminists who advocate that putting too much attention on getting women mammograms all the time ignores things like heart disease and domestic violence, which have a far greater impact on far more women, in favor of trendy pink things that are funded by certain multimillion-dollar agencies.  (Certainly, mammograms for all! Yes! And genetic testing!  But apparently the treatment after your diagnosis has not improved in its dignity, humanity, side effects, or possibly even effectiveness. Why are we making pink stand mixers and tape measures instead of fixing this?)

People who are used to critically evaluating science and how it gets reported won’t be surprised by anything in this article.  But far too few people have the skills and abilities and desire to do that.  Also, the article makes the good point that chasing money can corrupt anyone, which it does.  (Again, duh.  If your study is founded by a drug company, you tend to find that the drug works.  Duh!)

People don’t conduct good studies unless they have been specifically trained to do so.  Doctors don’t usually get this training.  Peer review has broken down when bad studies get published, but you have to ask who the peer reviewers are.   Many of the cognitive biases that scientists have, which are pointed out in the article, are normal biases that EVERYONE has.  Again, rigorous training in research methodology can make those less likely, but no person is immune from those biases (even those who study the biases)!  Ioannidis has his own biases, too, because he is a human being; I believe most of what he says, though.

Even the journalist of this article makes a common error here:  “(one large randomized controlled trial even proved that secret prayer by unknown parties can save the lives of heart-surgery patients, while another proved that secret prayer can harm them)“.
There is no such thing as PROVING a theory.  The word proof is misleading.  I do believe that the study found that secret prayer is associated with recovery from heart surgery; there are a lot of spurious correlations out there.  But finding SUPPORT for a hypothesis is *NOT* the same as *proving* it is “true”.  Again I say: science education, people!

Um, is that enough ranting for you?

#2 says:  Hm, I wonder if I should add this article to my stats final…  Usually the Atlantic has much higher quality science stuff.  The way I see it:  Correlation is not causation, but in order to have causation you first need to demonstrate correlation.  That’s what all those random results on food and exercise that come out of the Harvard nurses study do.  Second:  5 percent of the time you’re going to get things wrong with a 95% confidence bar.  And if you have a small sample size, you’re more likely to get an insignificant result even if there is a true effect (and 5 % of the time you get a bizarre result because you randomly grabbed some outliers).  Replication is important!


10 Responses to “In which I critique an article that critiques articles”

  1. Everyday Tips Says:

    There is nothing I hate more than reading an article that I think I will like and then something takes me out of it, like flawed thinking, or even a misspelling.

    I am with you on this one!

  2. Roshawn @ Watson Inc Says:

    Yes, this is very common. Even though I think we are rightfully laying a lot of blame on the media, I also think that the scientific training necessary to truly understand the limitations of research is beyond many people. I’m not arguing a lack of intellectual capacity as much as general disinterest (i.e. Jersey Shore more entertaining) and being unaware of the difficulties in interpreting research.

    I presented research at a conference a couple months ago. I discussed a positive association. I was surprised at how many people thought that this finding was ready to influence practice guidelines. In actuality, there are many additional studies necessary before it could be strongly considered for any guidelines.

    • nicoleandmaggie Says:

      The lack of general basic statistics knowledge in the population makes me cry.

      (Recently a student was in my office hours and said, “Gee, Dr. #2, a lot of stuff makes you cry, doesn’t it? Us not coming to office hours, not understanding that correlation isn’t causation, us getting worse than a B on the exam…”)

      • Money Reasons Says:

        I once knew a girl who was otherwise pretty intelligent, struggle with the concept of how to figure the 15 percent for a tip to give to the waiter. Later I learned that she couldn’t do percentages at all…

        I don’t look down on her though, some of us are stronger with logic and math than others. She a much better speaker than I am… To each their own :)

      • nicoleandmaggie Says:

        Percentages can be hard and math can be scary if you don’t have any confidence in it. She probably had a bad math teacher one year who broke her spirit. I get this all the time with tutoring. Usually I teach them one thing, like fractions or negative numbers, and it all becomes clear. I need to get back to tutoring sometime. Maybe when DC is older.

  3. David Freedman Says:

    You’re right that the media, public and many front-line physicians oversimplify, distort and misunderstand studies, and it’s true that the first passage you quote from my article refers to the headlines made by certain studies. But I fear you’ve misread the article. The trouble with the research described in the article applies directly to the study findings themselves as published in highly regarded research journals, not to the way they were interpreted by the media or anyone else. John Ioannidis, the subject of the profile, is a Stanford, Harvard and Tufts professor fairly revered by the medical research community, and he is the one saying that medical research is in terrible shape, and his highly cited research papers document how and why in great technical detail. All this is explained in my article. Perhaps you ought to take a look at Dr. Ioanndis’ papers, which are mentioned in the article. You can say that all we have to do is just ignore the bad studies and pay attention to the good ones, but when the good ones turn out according to scientists themselves to be a tiny and not easily identifiable fraction of the studies published even in top journals, I don’t know how you can claim that science is working just fine, and that any problems are all the media’s, the public’s and physicians’ fault! But I respect your opinion, of course, and I’m glad you took the time to discuss the article in your excellent blog.

    • nicoleandmaggie Says:

      I wasn’t clear, I fear. I was referring to the way the article itself was written being an example of “media interpreting science”. I have no particular quarrel with Ioanndis’ findings; I think they are probably right! And I did say that in there, I think. But I think that the finding that Ioanndis is correct (the finding that things get published in weird ways) is being portrayed a certain way in the article I linked to. Does that make more sense?

  4. eemusings Says:

    Ideally, it would be great if we had more journos with specific education/experience in science, reporting on the field.

    But we don’t.

  5. In which I critique — ah, the hell with it. « Grumpy rumblings of the untenured Says:

    […] have been known to critique science before.  Sometimes things on the internet make me want to critique bad […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: