Aiming the Obesity Blame Game at Working Moms

An old and reliable bit of clickbait has just surfaced from a rather obscure journal. In SSM – Population Health, Emla Fitzsimons and Benedetta Pongiglione claim to have “causal evidence” that a mother’s employment effects a child’s BMI. Naturally, British headline writers went nuts. So now we have another round of the obesity blame game aimed at working moms.

The Times won the contest for the most offensive version, getting straight to the point with its “lazy lardy” alliteration. But the real laziness here is intellectual. And we don’t mean the kids or the moms.

Strong Adverse Effects?

These researchers don’t stop with claiming to have the first evidence that a mom’s work causes childhood obesity. They go on to write that it is a “strong adverse effect.” Bold language. So we asked obesity researcher Michelle Cardel if this was justified. She told us:

No question that they are overstating the value of a fixed effects model for causal inference. On top of that, most of their “significant” relationships do not seem as strong as they are suggesting, based on the statistics in their paper.

Most of the correlation coefficients in this paper have p-values no better 0.10. Typically, the threshold for statistical significance is a p-value of 0.05 or smaller.

Assuming Causality from a Correlation

At the end of the day, these folks are simply documenting a correlation. Based on that, they’re saying they have evidence of causality. They are using a statistical method called a fixed effects model to give themselves more confidence in this leap of faith. But it’s still a leap of faith.

Professor and Dean David Allison explains:

I believe they have a stronger causal inference than an otherwise similar study that did not include household as a fixed effect. By doing so, they eliminated some confounders, but not necessarily all.

Twitter isn’t buying inflated claims from this study either. Obesity researcher Dr. Clare Llewellyn tweeted:

This is one of the most stigmatising reports about childhood obesity I have read in a long time. It’s appalling. No mention of the wider structural societal issues.

Without a doubt, all we’ve gotten from this study is absurd sensationalism. Obesity is complex to begin with. Add in all the complexities that come into play for working moms and things get really messy. So suggesting for a minute that this correlation proves that working moms are somehow causing obesity to rise in the U.K. is foolish. In fact, blaming parents for childhood obesity is cruel, destructive, and misleading – all at once.

Correlation and causation are very different things. Especially in complex systems.

This is an interesting study with interesting methods. But the authors have done everyone – including themselves – a disservice by overstating their claims.

Click here for the study and here for more perspective on the foolishness of the obesity blame game.

Immigration, Assimilation and the American Dream; photograph © Jeffrey Smith / flickr

Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.

March 11, 2019