Skip to content

“Sometimes research just has to start somewhere, and subject itself to criticism and potential improvement.”

Pointing to this careful news article by Monica Beyer, “Controversial study links pollution with bipolar, depression,” Mark Tuttle writes:

Sometimes potentially important things are hard, or even very hard.

Sometimes research just has to start somewhere, and subject itself to criticism and potential improvement.

I think this kind of thing supports our desire for high levels of transparency, difficult though that it may be in these circumstances.

I agree.


  1. jim says:

    Cool. I like that there is a companion paper that assesses the work in the appropriate “scientific analysis” tone. That’s a great approach. I can vaguely imagine this becoming a more widely used approach for controversial results – which could allow more controversial results to be published.

    But I still find of the language used for these types of studies a little bothersome. The fact that in a single study there is co-occurrence of two phenomena – does that imply they are “linked”? I’m not sure that’s a fair word to use. The word “Linked” initially implies a causal relationship, which is definitely not appropriate here. It could also be construed to imply that the two phenomena are linked through a common causal effect, although that’s probably not the first thing most people would think of. But they could also be completely independent. It could be, for example, that people with mental health issues are more likely to be poor and live in areas with low air quality because they are poor. Would that appropriately be called a “link” between air quality and mental health?

    • Jeff says:

      The headline seems fair to me. The paper’s abstract says it found that “air pollution is significantly associated with increased risk of psychiatric disorders,” so I read “links” in the article headline simply as “draws a connection between,” which was indeed the point of the paper.

      The causal language that jumped out at me in the article was “in this new study, the team looked more closely at how a specific environmental factor — air pollution — affects the brain and the likelihood of psychiatric disorders.” This seems false. The team *looked at* (observed) correlations and *thought about* mechanisms. You can debate whether the data support the hypotheses, as Ioannidis does, but this sentence distorts the thrust of the paper.

      Oddly, the article also says “Prof. Ioannidis eventually concludes that a ‘causal association of air pollution with mental [conditions] is an intriguing possibility,'” which is true only if “eventually concludes” means “states in the very first sentence.” This seems to me to soften the skepticism: “There are problems, but this is intriguing” sounds less damning than “this is intriguing, but there are problems.”

  2. David says:

    I think areas with worse air quality can be thought of as poorer areas. A link between economic depravation and Psychiatric Disorders may be more compelling. This research would be a good starting point, but there are other, more important factors that may contribute to mental illness.

    It is hard to take research between arbitrary factors at face value. First someone ought to reason that biologically it is possible that poor air quality can lead to mental illness before such subsequent research can come to be accepted.

    • Particulate pollution can be inhaled deep into the lungs, absorbed in the bloodstream, and cross the blood brain barrier. it’s entirely plausible for particulates to directly affect brain health. there are a number of indirect effects as well, such as through inflammatory responses and mental stress caused by poor health in general.

      • jim says:

        I agree this is plausible. I’ve also seen research suggesting that freeway noise can affect children’s development. I don’t recall the details of that research but the suggestion seems plausible, especially these days when the roar of freeways is so immense and continuous for 18-20 hours a day.

        What happens to the noise in the Big Dig? Is there a mechanism built to dampen the sound or does it just dissipate into the surrounding earth? I wonder how much traffic it takes for the sound energy to become meaningful in an engineering sense. A while back I was stopped at a construction zone and had the chance to get out and inspect the freeway surface. The fine concrete layer was completely worn off and the surface looked like a course conglomerate with a rough polish. Most of the freeways around here are like that – many actually have wheel ruts in the concrete. Lots of energy to wear that surface away and lots of noise doing it. Huh – now that I think about it the noise is probably much worse after the fine concrete surface wears off, because the rocks in the concrete are mostly much harder than the concrete itself. Poor maintenance leads to lower IQ?

      • David says:

        I would say this reasoning is not very convincing.

        The problem can be attacked differently, however. In the “Third World” countries, the air quality is much worse. I can say this because I live in one.

        However, there is no evidence that mental illness is more prevalent in these countries.

  3. david says:

    These are the best estimates I could find for Schizophrenia.

    About 7/1000 in the United States.

    In India it is about 3/1000.

Leave a Reply