Skip to main content.

Is Absence of Evidence Evidence of Absence? Why it Matters

Is Absence of Evidence Evidence of Absence? Why it Matters

Picture of Kate  Benson

A few weeks ago Slate writer Brian Palmer accused New York Times writer Jane Brody of using a red herring for a lede and promoting a theory that he believes is not factually substantiated. But, did he then do the same?

Mr. Palmer leads by lumping all patients with a specific disease in with other unrelated groups who are hot buttons for scientists and journalists alike – perhaps marginalizing them for many readers by saying pay no attention to these patients and their claims, they are just as wrong as these other groups.  This is becoming a divisive tactic used more and more often in science and journalism to invalidate the patient's lived experience - on which they are the experts. I can’t speak for scientists, but journalists can do better.

But, let’s blow the smoke screen of patients, advocates and their views aside and look at the journalism as well as the science Mr. Palmer cites.

Mr. Palmer attributes most of his “facts” to a nearly ten-year-old Letter to the Editor written by a source.  A decade is a long time in science and fresher data would have been better if available.  Mr. Palmer also ignores many of the reasonable scientific explanations for what is often termed chronic Lyme.  As Mr. Palmer wrote in 2012, not all tick borne diseases are known and or have a definitive test.  And as Beth Daley at the Boston Globe writes, the number of cases of Lyme disease is severely underestimated according to the Centers for Disease Control.

As well, misdiagnosis is a possibility as many clinical symptoms are common to many diseases. Another explanation could be that the immune system doesn’t switch off following the infection - a theory that could render the argument about spirochetes being present moot although there is conflicting evidence on this as well. And don’t forget possible reinfection and co-infections.

All in all, it is often difficult to say definitively in an underfunded research field and even when there is funding. Scientists knew about the link between smoking and disease long before they could scientifically prove it.  This may turn out to be the same – or not. I doubt either Mr. Palmer’s article or Ms. Brody’s piece will be the last word in the contested diseases wars in which patients often tend to be collateral damage.

Instead Mr. Palmer suggests that middle age has much in common with this disease.  Really?  Heaven save us all from middle age. Yes, there are symptoms involved that science does not objectively measure well or accurately, such as fatigue and pain, but does that necessarily render them imaginary, normal, minor or middle age? Sometimes science, like journalism, is a matter of asking the right questions.

Journalists generally use the lexicon carefully in order to mean what they say and specificity in science is much the same. The differential key here may be not just specificity, but severity as well according to researchers in other contested diseases.

Mr. Palmer advocates for objectively proven data saying arguments without such are “... like saying there are three possible explanations for an apple falling on your head, and one of them is gravity.”

But here’s the thing: the letter to the editor he cites claims chronic Lyme is a “functional somatic syndrome,” which is a recently revised term for psychosomatic but with no official standing. (A closely related branch of that ideological tree however, has caused quite a stir regarding the DSM 5 – with lack of empirical data being one of the issues.)  Nor is there an objective diagnostic test or objective data yet proving that functional somatic syndromes exist outside of a controversial theory held by a relatively small group of mainly psychiatrists.

Journalism is about asking the same hard questions of everyone. Giving one group a pass on hard evidence for example, but not another may be human, but it isn’t necessarily good journalism.

Here's the thing: scientists and science journalists often say correlation is not causation, but citing absence of evidence as proof of absence in science has similar problems as medical history shows. Not to belabor the obvious, but science rarely comes in an unchanging, neat and tidy package factually tied up with a perky bow on top.

I suspect Mr. Palmer was given the citation by a source he interviewed, but did not personally fact check the citation or interview experts who could point out the fallacies in the claims made in the cited letter.  Deadlines and thorough fact checking sometimes conflict and some might argue that you can only go so far.  What makes it relevant here is that Mr. Palmer accused one journalist of not providing factual evidence for the assertions in her story, and then appears to counter her assertions in part with other unproven assertions of his own.  (Another of Mr. Palmer’s original assertions was also corrected by Slate – the correction can be found at the end of the article.)

When it comes to the politics of disease it pays to be careful no matter whom you write for or what your topic is. Although the topic was different,Gary Switzer says in part, “journalists (can) become unwitting mouthpieces for incomplete, biased, and imbalanced news and information.”  

I would add that it isn’t necessarily bad per se to quote or cite what are in essence the scientific equivalents of “opinions,” but transparency is essential and the reasons they are opinions should be clearly clarified for readers. It's not enough to simply present conflicting "facts" and "opinions" - context for the controversy should also be required. 

Image by Marshall Astor via Flickr

Leave A Comment

Announcements

Do the competing bills put forward by House Speaker Nancy Pelosi and Sens. Charles Grassley and Ron Wyden have a chance of becoming law? This webinar will give an overview of the proposals and weigh in on the future of the battle to curb soaring drug prices. Sign-up here!

CONNECT WITH THE COMMUNITY

Follow Us

Facebook


Twitter

CHJ Icon
ReportingHealth