Salford’s Hot Food Near Schools Ban – A Nanny State Policy Without an Evidence Base

They justify their new SPD – restricting such uses within 400m of schools unless they only open after school hours

A 2008 report from the Nutrition Policy Unit of London Metropolitan University found that food outlets in close proximity to, and surrounding schools were an obstacle to secondary school children eating healthily. Both the Chartered Institute for Environmental Health

The problem is this research is totally discredited, and foillowing the recent windfarms case policies without supporting evidence are illeagal, I repeat from Jan on this blog

There is no dispute that public health is an important material planning consideration – the work of HUDU has been very important in raising, rightly, its profile. There is also no dispute that spatial proximity of one use to another – and the impact of one use on another is the very foundation of planning.

There are many issues relating to spatial proximity that are material however which will not besignificant enough to weigh in planning judgement, that is even before any national policy which determines the weight to be given to the application as opposed to other matters. For example proximity of a housing scheme to an SPA would me material in planning terms because of the likelihood of visitors, however the scheme might be far enough away as to not be significant. It is the significance of one marginal extra A5 use which is the issue.

In a densely built up urban area like Tower Hamlets there is likely already to be many many A5 outlets close to schools. Indeed a mapping exercise of fast-food outlets in Tower Hamlets, undertaken by Tower Hamlets in partnership with the Food Policy Unit at City University, found that there were 627 fast-food outlets, newsagents and groceries and that there were 42 junk food outlets (including fast-food restaurants and cold food outlets) per school in Tower Hamlets. So here we are talking about the marginal impact of the 43rd use, how big an impact will that have?

How far are these uses from schools? Tower Hamlets is so built up you hardly find any outlet which is more than 400m from a school, and so they apply a 200m threshold covering about half the Borough . An arbitrary figure which seems more policy based evidence (to avoid an unreasonable blanket ban) rather than evidence based policy.

So lets look at the evidence. The Tower Hamlets partnership commissioned a study from Public Health consultants Dr Foster together with Land Use Consultants.

There are four issues with any piece of research aiming to show geographical public health impacts sufficient to refuse a scheme:

  1. Is the result statistically significant;
  2. Is the strength of the effect significant;
  3. Would the effect of a restrictive policy have negative significant effects?

The first two issues are often confused. A result might be statistically significant because of a large sample size but the effect might be very low in strength (to the extent of not worth bothering about. I am troubled by the Dr Fosters/LUC study because it makes no attempt to prise out the different aspects of significance, all it is is a literature review with very unclear results.

For example an extract:

The same study that found associations between eating fast-food and a high BMI also found that proximity of fast-food restaurants to either home or work was not causally related either to eating at fast-food restaurants or to higher BMIs. (Marmot (2010) op. cit.; Cummins and Macintyre (2006) “Measuring outcomes from active transport interventions for children”, inInternational Journal of Epidemiology, 25, p100-104)

Research published only this year has found no correlation between the proximity of fast-food takeaways to schools and childhood obesity (although it should be noted that this was based solely on a ‘buffer zone’ of 800 metres, with no analysis apparent at the 400 or 200 metre level). (Currieet al. (2010) “The effect of fast-food restaurants on obesity and weight gain” inAmerican Economic Journal: Economic Policy, 2:3, p32-63)

Another American study found that, while proximity to food stores impacted on dietary intake and weight, proximity to restaurants (including fast-food outlets) did not. (Howard, Fitzpatrick and Fulfrost (2011) “Proximity of food retailers to schools and rates of overweight ninth grade students: an ecological study in California” in BMC Public Health, 11:68)

Another UK study found no independent relationship between indicators of healthier eating and the local retail environment. (Pearson et al. (2005) “Do ‘food deserts’ influence fruit and vegetable consumption? A cross-sectional study”, in Appetite, 45, pp195-197)

Now the literature review did include other studies which showed a small but statistically significant effect but the literature review overall is entirely unclear and not firm enough to base ANY planning policy changes on.

Furthermore a look at the research showing a small positive correlation between concentration of fast food outlets and obesity shows that it does not correct for the effects of spatial autocorrelation – I have a special filing cabinet for such studies as I have said many times, my waste paper basket. What do I mean by this? That the studies fail to correct for other factors which may cause obesity. For example dense urban areas are more likely to have concentrations of deprived ethnic minorities, the dietary habits of which may create risks of obesity – such as for example as is likely in Tower Hamlets a diet with large doses of clarified butter. Also densely developed urban areas have large amounts of schools which will automatically create a population which may be prone to obesity before any issue of fast food outlets arises.

But even if a correlation is shown the effect is likely to be very small. For example in Tower Hamlets only 1 in 5 uses a fast food takeaway more than once a week, so here we are talking about the marginal effect of the 43rd use on 1/5th of the population – very small.

If that extra 44th use which restricted what would children do, as likely as not they would simply walk a few extra metres to the next nearest use. As local competition would be restricted they might also pay more, and having to walk further may be late after lunch.

None of this is to suggest that obesity amongst children and teenagers is not important or that there is no correlation between proximity and consumption. This is not the kind of liberatarian rant you got at Mayor Bloomberg’s proposal to restrict go large coke sizes. Rather I am simply suggesting this policy falls way below the normal thresholds for incorporation of public health issues into regulatory policy.

If you really want to impose a policy which is shown to have positive public health outco0mes you make a product so expensive and so infrequently available that you control consumption. The classic model being state alchohol sale monopolies as in Sweden and India (and historically some Candadian states). There is wealth of evidence that this restricts consumption – though little on the difference between the price effect and the proximity effect. Diane Abbot hints at movement in this direction with her remarks on local shops sales. Do we really want to go down that illiberal route especially as it would harm greatly the viability of so many (ofrten ethic minority owned) small shops.

One thought on “Salford’s Hot Food Near Schools Ban – A Nanny State Policy Without an Evidence Base

  1. Pingback: Why I won’t be supporting the proposed fast food ban in Medway | No Sandals Required

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 )

Connecting to %s