The Grocer asks: The government made it clear in its White Paper that the FSA needed to work together with the industry to tackle obesity. Do you accept that this hasn't happened?

The FSA response: That's not true. When we set out to develop a tool to identify the foods that were high in fat, salt and sugar (HFSS), the first thing we did was to sit down with a small number of experts and agree a specification for the review of the available literature. We set up a small advisory group, with a mix of independent experts and experts nominated by the food and drink industry. Gaynor Bussell at the Food & Drink Federation and Karen Tonks at the BRC were part of that process from the very beginning.

But the FDF was only allowed to participate in the group on condition that it did not challenge the concept of nutrient profiling. And it resigned in protest at the direction of the system in development.

We are not just listening to manufacturers, but consumer groups and health professionals. We set out to consult with all parties. It was a long process, which involved many stakeholders, on both the advisory and consultation group. And it was trialled with 300 nutritionists.

So it was a very wide and engaging process. Hence our view that the model is robust and appropriately developed. And all participation in the development of the Nutrient Profiling Model was completely unconditional. We talk with the food companies all the time. One-to-one and via the trade associations. They have all said how pleased they are to be working with us.

This is not our impression. Suppliers have told us they will not criticise, on the record, any action by the FSA on obesity, because they are terrified that by speaking out, they will be singled out by the media and lobbyists and the FSA.

To say they are terrified is not the case. Why would they be frightened?

They should be prepared to stand up and have a proper dialogue with us. If they are not prepared to talk with us, that is a concern.

Why did you choose a model based on a 100g measure when we all know that some products, such as cheese or raisins, are simply not consumed in such gargantuan portions?

We went through a great many models and a great deal of consultation before we decided on the current one.

We looked at a 100g base, a portion base and an energy base. The advisory group settled on 100g because it worked best. These were debated within the advisory group and once the group was content it went out to public consultation. They concluded that the per 100g model worked perfectly well and that there was no need to look at something more complicated, ie a per portion approach.

But it doesn't work perfectly well, does it? It's seen a number of healthy and nutritious products, such as cheese, raisins, honey and bran flakes, banned from advertising to children

Your definition of what works is different from the definition that the academics were working to.

They were looking at the way the model categorises foods. This is an innovative piece of research. We were starting with a blank sheet of paper as there has never been anything done quite like this before and certainly not for this purpose.

We needed to find a more objective way to help Ofcom achieve a better balance in the advertising of food and drink to children. That is sometimes forgotten. People focus on the model without looking at the basis on which we are trying to do this.

So are you saying that it's perfect?

We have never said it's perfect. But we don't have any particular concerns about it.

We are saying that it's fit for this particular purpose. The Nutrient Profiling Model was developed for Ofcom. Should there be a need for another nutrient profile we would have to go back to the drawing board and look at what it is for, and what's out there.

Nutrient profiling is not about banning the advertising of cheese and olive oil. It was designed to help Ofcom rebalance the proportion of advertising of HFSS foods to children, on television. With hindsight it has not helped people's understanding that we started [to develop] traffic lights and the nutrient profiling model [at roughly the same time].

Are you not concerned that healthy and nutritious products, when taken in the context of a healthy and nutritious diet, are effectively banned from advertising on children's television? For example, do we not want children to eat cheese?

The dairy category was an important one in the development of the Nutrient Profiling Model and experts were always clear that there must be dairy products advertised to children, so any model which didn't allow some to be on the right side of the line would not be considered appropriate. We have a model with some dairy products on one side and some on the other, and they were content with that.

Why doesn't the Nutrient Profile Model consider more positive nutrients?

We considered calcium and iron but we had a lot of response from industry asking us not to include them because they didn't have the information on the amounts present in many foods, and explaining that it would be extremely costly to analyse these minerals and nutrients.

So in phase two of the development process for the model, we examined the possibility of a proxy for calcium and iron that had no impact on cost.

We used protein [as the proxy] because food that contains calcium and iron also [typically] contributes protein.

Are you not concerned that, because of the limited range of nutrients that the model considers, food manufacturers will reformulate using artificial sweeteners and additives and preservatives?

When we worked with industry to cut salt, you would think they might have put salt replacers back in but that wasn't the case.

Manufacturers and retailers are conscious that people want foods that are natural. Therefore manufacturers are looking at how to work within these parameters.

But it is important not to look at the model in isolation. We don't and neither do people we talk to in the industry.

We have got a number of initiatives such as our work to reduce salt, and we are developing an approach for saturated fat and calorie intake, using energy density in particular. These are the onuses on businesses in terms of the reformulation of products.

But the model is often helpful in terms of reformulation. If manufacturers are going to reduce saturated fat, for example, they have got to put something else into the product.

That can mean improving a product. So we very often find there isn't a conflict between these [two objectives] at all.

There has been a tremendous response from manufacturers and retailers in terms of reformulating products and getting healthier eating on the agenda and we welcome that.

Three years ago we didn't have that level of engagement.

The Grocer has identified a number of alternative schemes that have been developed, or are in development? Have you looked at them, and if not, will you do so?

Some of the options you have talked about were developed in-house by manufacturers, and have not been subject to any serious public scrutiny or scientific assessment and we can't afford the luxury of working on preference, as we are being held to account, and rightly so. But for pity's sake why don't they come and talk to us?

We do not know what a junk food is. What we can tell using the Nutrient Profiling Model is what food is high in fat, salt and sugar, and we know you must be careful of how much and how often you eat those foods. That's what this model is trying to do.

Is the Food Standards Agency doing enough to encourage exercise among young people?

It's a really critical issue. The role of diet is part of a wider healthy lifestyle, but the agency's remit is the food part of that equation. We cannot regulate jogging or exercise.

But you can talk about portion control. Why does none of your messaging mention temperance and restraint? After all, most people who are obese eat too much, don't they?

We are clear in all our messaging that diet is seen as part of a healthy lifestyle.

What's your reaction to products that have reformulated but have still fallen foul of the ban?

We would question whether a product that is still high in fat, salt and sugar is something they feel they should be promoting to children. You single out products, like manuka honey, that are not advertised to children.

Why should a precedent be set that prevents manuka honey manufacturers from advertising on television in the future?

We are concerned here with the food that's advertised to children on television today.

You may think that's splitting hairs but the fundamental point is that we want a better balance in children's diets, and the model is a way of helping Ofcom decide that.

If levels of obesity haven't changed in a year, then what?

We are yet to decide the exact timing for review.

To actually measure the impact on obesity is stretching the purpose of the review, however. It will be looking at the impact of what foods are advertised to children.

We are committed to a review of the Nutrient Profiling Model in a year but are always interested to hear what other experiences people have, and for them to talk to us about it.

Are you doing anything else, beside the development of the Nutrient Profiling Model to reduce obesity in children?

We have several cooking buses now which are the most wonderful Tardis-like things that develop into complete kitchens and they travel around the country and take children from schools and work with the children and work with the teachers.

Given that 60% of your remit is about the regulation of food consumed outside the home, do you have any plans in this area?

We don't envisage a traffic-light labelling approach to restaurants but we do very much encourage restaurants to provide information about their food.

I remember doing some research a few years ago and consumers said 'when we are going out to dinner, we know it's a treat and we don't want to be bothered with [fat, salt and sugar levels]'.

But I think that may be changing because so many people eat out so much more. But there are some quite good examples of restaurants providing much more labelling themselves and that's good, and we just need to make sure it's very obvious and not hidden away in the corner.

What are you doing to encourage children to eat positively nutritious foods?

I suppose you could say that putting 'green' signs on [labels] is a positive attribute. What we need to find out is what helps people.

I heard a woman in a focus group for Waitrose who said: 'I stick to green and ambers during the week, and on Friday I allow myself three reds.'

I don't think the public is viewing this in the way you describe - we have no evidence of the public saying 'red food means something you must never, ever eat'.

There is one company, which I will not name, which has been in discussion with us [about its indulgent products].

It is fascinating as they know perfectly well that every segment is going to be 'red'

They are quite happy with that because they want to position their product as a treat.

But equally you've got McCain chips who have produced a very low fat salt-free chip, who are able to put a complete green on every category in relation to their chips - that's a fantastic marketing tool for them. Here are the healthiest chips in the world.

So it works in many different ways. It's not quite as straightforward as the way you describe it.

What chance is there that you will review the Nutrient Profiling Model? Is it the case that if a better model comes along, you would consider it as an alternative?

We said right at the start of this process that we will review the model after a year of use.

That's where we sit at the moment. But it's an innovative process. With any new invention you can always introduce improvements.

There is nothing else on the market that has gone through this process of development and consultation and peer-led review. It's about reviewing its purpose against TV advertising to children.

In a year we will look at it. If the approach is still the best way of categorising foods for this particular purpose we will continue. And if, in the meantime, other ideas come through that have gone through a similarly robust process, we will welcome that, but so far we haven't seen that happen.

It's very important that we review everything we do.

But we have got to give it time to see whether there has been any shift in the advertising of products to children and which way it has

occurred. n

This interview is based on a meeting with senior members of the Food Standards Agency on 13 March 2007. Further questions were posed to Dame Deirdre Hutton, the chair of the agency, at the IFE VIP address earlier this week.Ready meals are considered as a whole. Why don't you take the milk into account when you consider breakfast cereal?

I'd put that question straight back to the industry. There are cereal manufacturers that state just the dry weight of the cereal, and there are those that give the nutritional information with the milk included. They can't decide which way to do it so it is confusing for consumers. If you look at EU legislation, everything is done on a 100g as sold basis. So I'd put the question straight back to them and ask why don't they sort out their portion sizes and whether cereal should be consumed with milk or fruit juice. When it comes to cereals it's very challenging because you have different weights of cereal so the weight will vary. Some will take a very heavy weighted product and count it with milk, some won't. Until the industry can sort it out themselves it's very difficult.You are vice chair of the EFSA, which is developing a nutrient profile for labelling claims. Are you lobbying for the EFSA to use your model?

Absolutely not. The FSA has never

lobbied the EFSA on the subject of nutrient profiling. Our model was developed exclusively for Ofcom. It will not be used for any other purpose, not even other forms of advertising.Sir John Krebs is a scientist. You're from a consumer background. Is the FSA now science or consumer-based?

There are more strings to my bow than you suggest. I'm on the board of the Financial Services Authority and worked on the Curry Commission, the Foresight Programme and with the Food Chain Centre. But 40% of our staff are scientists and we are advised by nine independent scientific committees.