Wednesday 30 November 2022

Great success! Mexico's sugar tax

There's some terrific cognitive dissonance in this article in Frontiers in Public Health. You may fondly recall the Mexican sugar tax which supposedly reduced sugary drink consumption by 6 per cent or 12 per cent, depending on who you get your news from (it actually reduced it by about 3 per cent). It was much talked about in the months before George Osborne inflicted a sugar tax on the UK.

Did this 'world-leading' policy lead to a reduction in obesity in Mexico? You might think so according to the authors of 'Childhood obesity in Mexico: Influencing factors and prevention strategies'. They say... 
... given that sugar-sweetened beverages are well understood as a casual factor of overweight and obesity, the year 2014 saw the implementation of a tax of one Mexican peso per liter on beverages with added sugars, as well as on food items with an energy content ≥275 calories per 100 g.

Alas, their figures tell us otherwise. 14.7% of Mexican kids were classed as obese in 2006 and there was "a sustained increase over the following survey years until reaching nearly 20.0% in 2021."

Yes, it's another big 'public health' win! The obesity rate amongst adults has also risen - from 32 per cent to 36 per cent since 2012. On the face of it, the widely lauded sugar tax of 2014 has made not a jot of difference to the thing it was supposed to tackle. 
It would be interesting to compare these figures to countries in central America which do not have sugar taxes, but idea of using a control might be a bit too 'sciency' for the authors who seem to have a pretty weak grasp of how to work with statistics. When they look for correlations between diet and obesity they say:

Snack foods, sweets and desserts, and sugar-sweetened beverages were not statistically significant contributors in the model (p = 0.112 and 0.098, respectively), but seem to show weak associations.

There is either an association or there isn't. Statistical significance exists to help us establish if they is a likely association. There isn't one here but the authors proceed to act as if there is.
The pseudo-panels of the study showed that those food groups considered as protective, such as fruits and vegetables, protected against overweight and obesity in school-age children. On the other hand, intake of snack foods, sweets and desserts, and sugar-sweetened beverages were associated with the presence of both outcomes.
No, they weren't.
They do it again with fibre.
We also found that greater fiber consumption was associated with lower overweight and obesity in school-age children, even when this association was not statistically significant.

How do this stuff get through peer review?
Although this association is inconclusive in the scientific literature, we suggest that a greater fiber content in food items indicates a lower caloric density, as well as a lower consumption rate and possibly a greater sense of satiety.
In other words, "we didn't find an association between fibre intake and obesity, and the rest of the literature is inconclusive, but here's what we reckon..."
Notably, when facing a problem of the magnitude of school-age child obesity, permanent government strategies and actions are required to ensure the containment, prevention, and management of this outcome in order to avoid serious consequences for long-term health and national development. To the present, various strategies have been implemented in Mexico, as well as in other countries around the world.
And none of them have worked.
Although these have been associated with some positive results, they are not public policy actions which demonstrate an overall effective impact.
That's one way to describe abject failure.
An example of this is the establishment of regulations on the sale and distribution of foods and drinks in the school environment dating back to 2010, and which were only put in place in 2012 and whose implementation was monitored only starting in 2015.
That didn't work either.
Other actions implemented on the national stage and not only targeting the school-age population include the 2014 tax of one Mexican peso on sugar-sweetened beverages, and one previous study found a reduction of 6% in buying of taxed beverages in 2014. Furthermore, from 2014 to 2015 changes were observed in the buying of taxed and non-taxed beverages, where buying of taxed beverages diminished by 5.5% in 2014 and 9.7% in 2015: an average reduction of 7.6% over the study period.

Did it lead to a reduction in obesity, or even a slowing of the increase of obesity, as was promised? As we have seen, it did not. 
Nonetheless, the tax is too small to expect the provocation of biological impacts such as a reduction in obesity.
Oh, so now you tell us. Didn't say that at the time, did you? If you'd said it wasn't going to reduce obesity, the government would have never implemented it. All the modelling, as recently as 2020(!), said it was going to reduce the weight of children. Now you tell us the tax was too small. 
Why is anyone still listening to these charlatans?

No comments: