Friday, October 28, 2011

In search of Scientific Journalism

There are very few cases out there of science in the media which I can actually take as reliable.  Those of you out there who have read some of my earlier rants know how badly I wanted to eviscerate the fools who wrote about DCA and cancer treatment.  There is a general trend I find when it comes to science stories in the mainstream media.  They tend to be obscene scare stories, or stories of outrageous new breakthroughs.  Occasionally, you get something about strange occurrences or weird anomalies, but they are pretty silent entries on the back pages and nobody ever really remembers them.  The stuff that makes the front page are either stories that say how terrifying something is, or stories that say how amazingly wonderful something is.

Your iPhone could be giving you brain cancer!  But acai berry cures Alzheimer's!  The Large Hadron Collider will cause a black hole to form in the center of the Earth...  and they predict that it will happen on December 21, 2012.  Make sure to stock up on chocolate and red wine, because they will prevent all illnesses with their "essential" flavonols.  Don't you have a flavonol deficiency?  Well, it shouldn't bother you anyway, because every vaccine you take is full of poison!

It amazes me at times when you see a news program where someone will have a doctor interviewed, who has the benefit of decades of research, large-scale data, longitudinal studies, and scientific development on his side...  and then they will ensure that his time on air is shared with some horseshit peddling activist whose knowledge of medicine lies somewhere in the realm of a tapeworm's understanding of quantum mechanics...  or maybe even as poor as Michele Bachmann's grasp of anything that happens to actually be true.

In the realm of scary stories, there are always cases where some journalist takes a simple little portion of a case study, usually not even data, but just a sentence or two discussing some part of the results, and taking that to mean the whole content of the study (when in fact, it rarely matches what the people were actually trying to say).  Even the long-since discredited paper by Andrew Wakefield which sparked the whole MMR scare was quite clear in stating that its results were operating on too small a population (only 12 children) to draw any sort of conclusion.  But the media at the time took it to mean something much more significant than even the original authors were willing to take it, regardless of any other deceit which underlied the data.  Of course, other follow-up studies had been done -- the largest of which involved over 500,000 children (by the CDC and Danish Medical Research Foundation), and showed precisely zero correlation whatsoever.  But do you see anything about that on CNN or the New York Times?  Now, of course, you can't just collect 500,000 data points overnight, and really showing the incidence rates involves actually following the subjects over a longer term (in this case, 7 years).  But that's how science works.  And it does work.  However, it only takes one idiot/Playboy bunny to pick up on a poorly written report of a useless paper (for the sake of argument, assume that we're still back in the time when the deceit in Wakefield's study was not yet exposed) and you have the formation of a new religion.  And like any other religion, absolutely no amount of fact, reason, or evidence can change their minds.

Media, of course, try to act in the name of balance by letting someone like Jenny McCarthy voice her nonsense on the air just as much as they'd like to let people who actually know something voice their statements...  and in some cases, you have people like Oprah Winfrey who feel it is far better to let the quacks, crackpots, lunatics, and insufferable morons have the majority if not the only voice in the room rather than let someone on who does all that science-y stuff.  It is not entirely out of a complete and utter disregard for science per se, but out of a complete and utter inability to comprehend the differentiation between facts, evidence, and opinion.  Journalists produce science news as if there is fundamentally no difference between a hypothesis based on large quantities of evidence and data collection and organization and analysis, and a random baloney you thought up based on the fact that the neighbor's kid started spinning things recently.

On the other end are the miracle stories.  I previously mentioned the fiasco over DCA, wherein somebody made a big deal over a study that showed relatively positive results on samples of living tissue.  But that is not the same thing as doing actual trials on patients.  You'd think a "science correspondent" would get that...  but no.  Some years ago, there was a small-scale study that appeared to associate slightly elevated risk of heart disease with ibuprofen usage...  but it was too small a population to make any sort of conclusion, and the degree of increase was so small as to fall within the error bars.  This much was made clear by the study itself.  But journalists are more twisted than that...  they decided to try an use another measure such as "relative risk"...  Relative to what, you may ask?  Don't ask such silly questions.  Because a journalist will just make it relative to anything that sufficiently inflates the figures.  So instead of a 0.02% risk over placebo, somehow it's a 35% higher risk than statin drugs!  But that's entirely antithetical to actual scientific studies since takers of statin drugs are hardly a valid control group.

But even aside from that, if you actually read the original sources, you'll find that no scientist ever says that small studies, individual inklings of correlation, tests done on animals or tissue samples...  are at all big news.  They are big from the standpoint of being able to merit further research and further study and development, but that only shows the job is not done at that point.  It's journalists who say it's a big deal.

Most any time I see an article begin with the statement "scientists have found", I'm not that likely to believe that any part of that statement is actually true.  I especially don't believe it for a second if some article in a normal press outlet says that scientists have found "a link" between something and something else.  When articles in the mainstream press say this, what it usually really means is that some study shows that there may be some degree of correlation which may or may not even be statistically significant.  Finding a "link" implies some sort of causal connection has been established, which all but never happens in any one particular study.  It tends only to happen in compilations of several studies, all of which explored and tested a particular hypothesis.  Journalists can get away with this white lie because most people, even those who know a thing or two about science, aren't likely to actually read the original sources.  Many times, also, the original sources might not even be available for free.  Most people who don't really know much at all about science will swallow it wholesale because they don't really grasp the difference, nor do they have any sort of ability to process why it is that finding antioxidants in chocolate is not the same thing as saying chocolate is now health food.  My condolences to women the world over...

Now I know there are exceptions out there.  People like Michael Specter, Peter Hadfield, Sanjay Gupta, etc. are all quite careful about how they report scientific data and reports and results.  Real science is after all, much more tempered than that, and involves a great deal more time to arrive at major developments.  That is the result of the careful and methodical and cautious nature of how science develops our understanding of the world.  The problem is that a very distorted picture of that is what is reported to the populace at large.  Exceptions being rare, journalists who cover scientific stories are charged much like any other journalist with the job of bringing something to the table that sells, and they need to put it in words that make the layman take notice.  Moreover, it has to go past subs, section editors, checkers, and so on...  and if they have no understanding in science and what actual scientific studies tend to say, then they too will not balk at serious corruptions of what many studies actually do say.  When it passes around being interpreted by idiots, you eventually get someone like Glenn Beck, who is about as unscientific as it gets, interpreting poorly reported information about otherwise good science, who will take that interpretation to radio and TV air time... whereby he can advocate that people self-medicate with dangerous untested drugs.

And yet somehow, scientists are the ones not to be trusted.

If you think this way, then I have a bridge I'd like to sell you...  on the condition that you go jump off of it after purchase.