Overselling the Microbiome Award for @nytimes on thumb sucking, nail biting protecting from allergy

I am continually torn about handing out "overselling the microbiome" awards to many "stories" that are coming out recently on new scientific studies.  On the one hand, many of these studies are quite interesting.  On the other hand, a huge number of them oversell the implications of the work.  And for some reason it seems to me that studies that could indicate a positive role for microbes in some way seem to end up with more misrepresentation than other types of work.  Mind you, I truly believe the cloud of microbes living in and on various plants and animals are likely to play fundamental roles in all sorts of important functions.  But my thinking this and my thinking it is likely does not mean we should go around overstating the implications of work in this area.

And that brings me to the latest example of such overselling  ... a story about thumb sucking and nail biting as covered in the New York Times: Thumb Suckers and Nail Biters May Develop Fewer Allergies

The science here is interesting  - it is based on a new paper testing for associations between thumb sucking and nail biting on the one hand and atopic sensitization, asthma and hay fever on the other.  The paper found the following: Children who suck their thumbs or bite their nails are less likely to have atopic sensitization in childhood and adulthood.

Interesting.  But a key part of this is that they discovered a correlation.  Lots and lots and lots of possible explanations for this correlation including some examined in the paper but none of which have been proven.  Some news reports do a good job of covering the topic and discussing how this is still just a correlative observation.  For example see this Washington Post article by Lateshia Beachum.  There they report on the authors comments where the authors seem to think this supports the hygiene hypothesis

“The findings support the ‘hygiene hypothesis,’ which suggests that being exposed to microbes as a child reduces your risk of developing allergies,” Hancox said in a statement.

But then immediately this is countered by some more careful thoughts

Hirsh Komarow, a staff clinician at the National Institute of Allergy and Infectious Diseases, isn’t entirely convinced about the study’s conclusions. “It’s an interesting observation, but it needs more analysis,” Komarow said.

And then he is further quoted with other possible explanations

Komarow also suggested that thumb-sucking and nail-biting could be indicative behaviors that either thwart or encourage allergic reactions. He said being part of a large family and being exposed to microbes from many siblings may affect a child’s allergic sensitization.

And there are other articles out there with a decent amount of caveating.  But sadly the New York Times article by Perri Klass is not so tempered. Here are some of what I consider to be statements without enough caveating or countering:

A new study suggests that those habits in children ages 5 to 11 may indeed increase exposure to microbes, but that that may not be all bad.

No no no.  The new study did not suggest that.  The new study is consistent with that, but it is consistent with many other explanations.

And then there is this:

These differences could not be explained by other factors that are associated with allergic risk. The researchers controlled for pets, parents with allergies, breast-feeding, socioeconomic status and more. But though the former thumb-suckers and nail-biters were less likely to show allergic sensitization, there was no significant difference in their likelihood of having asthma or hay fever.

Well it is nice that the authors of the paper tested for some other possible explanations.  But it is a giant and inappropriate leap to go from that to "These differences could not be explained by other factors that are associated with allergic risk"

And then there are multiple quotes from the authors which are not really caveated enough or at all

Robert J. Hancox, one of the authors of the study, is an associate professor in the Department of Preventive and Social Medicine at Dunedin School of Medicine, a department that is particularly oriented toward the study of diseases’ causes and risk factors. He said in an email, “The hygiene hypothesis is interesting because it suggests that lifestyle factors may be responsible for the rise in allergic diseases in recent decades. Obviously hygiene has very many benefits, but perhaps this is a downside. The hygiene hypothesis is still unproven and controversial, but this is another piece of evidence that it could be true.”

Some caveats here but not enough.  This is not "evidence that it could be true" but rather it is data consistent with that model, but also consistent with other models that have nothing to do with the hygiene hypothesis.

And then there is this:

Malcolm Sears, one of the authors of the paper, a professor of medicine at McMaster University in Hamilton, Ontario, who was the original leader for the asthma allergy component of the New Zealand study, said, “Early exposure in many areas is looking as if it’s more protective than hazardous, and I think we’ve just added one more interesting piece to that information.”

No this study did not show that early exposure from thumb sucking or nail biting has any protective benefit.  It showed a correlation between thumb sucking/nail biting and lower risk of sensitization.  It did not show any causal connection.  And even if a causal connection were found, one would still have to test for what was the mechanism and the mechanism could be many things unconnected to microbial exposure.

And then there is this

Dr. Hancox pointed out that the study does not show any mechanism to account for the association. “Even if we assume that the protective effect is due to exposure to microbial organisms, we don’t know which organisms are beneficial or how they actually influence immune function in this way.”

Yes, this is good in some ways.  But why would we assume this?  Stating this without caveats makes it seem like we should assume this.

and

Dr. Sears said, “My excitement is not so much that sucking your thumb is good as that it shows the power of a longitudinal study.” (A longitudinal study is one that gathers data from the same subjects repeatedly over a period of time.) And in fact, as researchers tease out the complex ramifications of childhood exposures, it’s intriguing to look at long-term associations between childhood behavior and adult immune function, by watching what happens over decades. 

None of these quotes are really caveating the claims.  And then the article ends with a statement that seems to indicate that this is all a proven fact

So perhaps the results of this study help us look at these habits with slightly different eyes, as pieces of a complicated lifelong relationship between children and the environments they sample as they grow, which shape their health and their physiology in lasting ways.

Yes, this study is interesting.  And yes, it might be indicative of a causative connection between exposure to microbes on thumbs and nails and reducing risk to allergy.  But no, the study did not show that there is a causative connection, just a correlation.  And thus we cannot conclude at this point that we should "look at these habits with slightly different eyes, as pieces of a complicated lifelong relationship between children and the environments."  The correlation could be due to other factors that have nothing to do with these habits.  And this is just a massive difference.  Shame on the New York Times for not reporting on this carefully enough.

And thus I am awarding a coveted Overselling the Microbiome Award to the New York Times and Perri Klass.

Hat tip to Mark Sagoff for pointing me to the NY Times article.