Null results in largest environmental study of its kind

The largest meta-analysis in the world of maternal exposures to air pollution and newborn birthweights, which combined effect estimates on about 3 million births from 14 international  research centers, found no link. That’s not what is being reported, however.

Air pollution increases chances of low birth-weight babies

Sooty air pollution in towns and cities increases the chances of women giving birth to small babies, new research has shown. A study involving millions of births around the world found that higher pollution levels raised the risk of low birth-weight. Although small, the effect is said to be statistically significant. At national population scales it could have an important impact on child health, said the researchers.

They face an increased risk of dying in infancy, as well as chronic poor health and impaired mental development. The new study, the largest of its kind ever conducted, focused on tiny sooty carbon particles called PM10s and even smaller PM2.5s which are known to be linked to heart and lung problems and early death….

As usual, media takes its lead from a press release. The University of California, San Francisco, press release headlined: “Maternal exposure to outdoor air pollution associated with low birth weights worldwide,” and touted this EPA-funded study as evidence of the benefits of reducing air pollution at all costs and “a lesson that all nations can learn from.”

The study, published in the online journal, Environmental Health Perspectives, (a regular source of poor quality environmental research), was equally misleading, concluding: “We found that maternal exposures to PM10 and PM2.5 during the entire pregnancy were positively associated with term LBW.”

Except the authors and the journal failed to admit that those “positive associations” weren’t even tenable and beyond random chance or modeling error. The combined estimated odds ratios were 1.03 and 1.10, respectively.

3% and 10% is well below the 200-300% that credible epidemiologists require to show a tenable link. In any reputable scientific journal, these would be null findings.

One response to “Null results in largest environmental study of its kind

  1. I have always been distrusting of ‘meta-analyses’ – adding apples to oranges and producing an estimate for a fruit salad recipe. One of the Corrolaries of Murphy’s Law states that “Nature sides with the hidden flaw,” so looking for needles in aggregate haystacks is inherently suspect.
    Nonetheless, it is remarkable that a ‘meta-analysis’ goes looking for a pre-determined reportable result and the result they find is null. It means that none of the individual studies were separately able to detect a reportable result.
    A null result in an individual study is rarely remarkable, but when it repeatedly arises in a large number of studies it is noteworthy.

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s