Is the Health Minister right about a link between deprivation and obesity?

Public Health Minister Anna Soubry caused a splash this week by apparently claiming, as The Times put it, that 'you can spot poor kids because they're fat'.

Speaking to the Daily Telegraph, Ms Soubry said:

"When I walk around my constituency you can almost tell somebody's background by their weight. Obviously not everybody who is overweight comes from a deprived background but that is where the propensity lies."

In particular, the Minister took issue with the diets of poorer children, who she said were being plyed with an "abundance of bad food".

So is there a clear link between deprivation and larger waistlines among children?

Certainly if you look at the prevalence of childhood obesity in schools based in more deprived areas (as measured by the Index of Multiple Deprivation) and compare it to those schools based in more affluent locales, there is some evidence that would support Ms Soubry.

According to the National Child Measurement Programme, approximately 13.7% of kids in their final year of primary schools in the least deprived 10% of neighbourhoods are obese, while almost a quarter of children of the same age are considered obese in the 10% most deprived schools. Looking at children just starting school there is a similar pattern:

However if we take the Health Minister at her word, she wasn't just talking about children considered obese, but also those that are overweight.

Here the data is less clear cut. While it does show that a slightly higher proportion of children attending schools in the most deprived areas are overweight than those in the least deprived, the overall trend is largely flat, and actually peaks in the middle deciles.

There are limits to what we can say based on this data however. It can only tell us, for example, how deprived the area in which each school is situated is, not how deprived the backgrounds of overweight and obese children are.

We should also be cautious about how we understand this data: while there does seem to be a correlation between the prevalence of childhood obesity and how 'deprived' the school's locale is, we shouldn't assume that one necessarily leads to the other. There could be a variety of other factors in each area that are contributing to the trend.

However the BBC has also taken a look at the issue, using the parental incomes of children as a proxy for how 'deprived' they are, and has found broadly similar trends.

So while the topic is undoubtedly a complicated one, there does seem to be some evidence behind Anna Soubry's claim, so long as we refer specifically to obesity.