A bad time for a bad obesity prediction

All you have to do to produce an obesity forecast that will be published in a medical journal is draw a straight line through the recent past and push it forward a few decades. That is literally all there is to it and it is why such forecasts are very likely to be wrong. 

Sure enough, they are always wrong. According to past predictions, the UK's adult obesity rate should have been 33% in 2012 (it was 25%) and it should be 40% by 2030 (seems unlikely; the figure for 2022 was 29%).

Predictions for the US are even more dramatic. To be fair, obesity is significantly higher there (41%). But the prediction that "by 2048, all American adults would become overweight or obese, while black women will reach that state by 2034" remains as implausible as it did in 2008.

Nevertheless, the media lap these forecasts up and the studies that contain them invariably urge policy-makers to regulate people's diets to avoid these awful outcomes. It's never too late!

The latest effort was published in the Lancet this month. Minor aristocrat and major killjoy Jim Bethell was all over it.

This is very powerful.
Most striking, USA obesity rates are accelerating!
Our current efforts are a total failure.
I’d guess UK data similar. Anyone know for sure?
For sure, to bend the curve we need a new approach. https://t.co/sy5yEwqZ1l

— Lord Bethell (@JimBethell) November 20, 2024

The study reckons that...

 

The projected prevalence of overweight and obesity in 2050 is estimated to be 81·1% (77·9–84·5) in adult males and 82·1% (76·7–85·7) in adult females. The prevalence of obesity is projected to increase at a more rapid rate than overweight, and faster among males than females. In males, the percentage change in the prevalence of obesity from 2021 to 2050 is estimated to be 32·8% (15·7–48·0), with a prevalence of 55·3% (47·7–61·8) in 2050. For females, the percentage change in the prevalence of obesity from 2021 to 2050 is estimated to be 28·6% (13·5–39·3), with a prevalence of 58·8% (51·1–64·1) in 2050.

 

In West Virginia and Kentucky, the authors reckon at least two-thirds of the adult population will be obese by 2050! 

 

 

All this is based on the assumption that "current trends" will continue. But if you've been reading the news recently you may know that current trends have already stopped continuing. As James Burn-Murdoch reported in the FT last month...

Around the world, obesity rates have been stubbornly climbing for decades, if anything accelerating in recent years. But now newly released data finds that the US adult obesity rate fell by around two percentage points between 2020 and 2023.

 
What makes this all the more remarkable is the contrast in mechanisms behind the respective declines in smoking and obesity. The former was eventually achieved through decades of campaigning, public health warnings, tax incentives and bans. With obesity, a single pharmaceutical innovation has done what those same methods have repeatedly failed to do.

  

I guess this news came out too late for the Lancet to halt publication. Awkward.

 

 

Burn-Murdoch is confident that weight-loss drugs explain the decline in the USA. I think we need to wait another year or two to be sure. Someone on Twitter mentioned the possibility that the decline is because Covid killed off a disproportionate number of obese Americans. It's an interesting thought. After all, it should have. And yet we saw no reduction in obesity in the UK (the 2022 figure is an all time high) and I don't think any other country has seen a fall in obesity rates. Burn-Murdoch also says that the fall in obesity was largest among the better educated, which fits the drug theory but not the Covid theory.

Let's wait and see.