Note of meeting: independent monitoring and assessment of front of pack signposting schemes

Rosemary Hignett - FSA ND
Claire Boville - FSA ND
Jill Pitt - NCC
Tim Marsh - NHF
Chris Werman - FDF (Kelloggs)
Kevin Hawkins - BRC
Linda Main - BDA
Moira Howie - Waitrose
Matt Carden - FSA ND
Karen Tonks - Tesco
Sue Davies - Which?
Judith Batchelar - Sainsbury's
Gaynor Bussell - FDF
Mark Read - FDF (Masterfoods)
Jennifer Poulter - Nutritionist
Mairi Hetherington - COI
Teresa Gilbert - IGD

Rosemary opened by briefly describing the background to the meeting. She explained that the purpose of the meeting was to discuss how to independently evaluate the various front of pack nutritional signpost schemes being used in the UK marketplace. It was hoped that the scope of the study could be mutually agreed with all participants to find out what works in practice for consumers and that all participants would be expected to accept the results of the independent study. She stressed that the exercise should be viewed as a learning opportunity and not a competition between schemes.

It was agreed that the research should focus on addressing the following question:

• In practice does the scheme help UK consumers, particularly those in C2DE socio-economic groups, make healthier choices when buying pre-packaged food?

and that the study should also address the following sub questions for each scheme:

• Do consumers notice the scheme?
• Do they understand it?
• Can they use it easily and quickly?
• Do they use it to make healthier choices?

There was some discussion on whether the study should target consumer groups identified according to incidence of obesity / diet-related health problems or ability / willingness to use label information as well as the sort of data that should be collected. It was felt that the terms 'at a glance' and 'quickly' would need to be defined in the research brief. The group agreed that the research should:

• focus on C2s, Ds and Es but should not ignore other groups
• establish whether consumers can actually use the various schemes
• look at what they do in practice
• look at choices both within and between food categories
• consider the context of whole-day choices rather than individual meal choices
• consider diet quality rather than calorie intake
• assess awareness and usage of schemes
• take account of Department of Health work on consumer behaviour

The BRC agreed to locate relevant research carried out by the Scottish Executive into socio-economic group purchasing behaviour and provide a copy to the FSA for circulation to all attendees.
Action: Kevin Hawkins

The group discussed the pros and cons of using a number of research methodologies:

Sales data - it was felt that data for a period pre- and post- introduction of signposting should be assessed. Retailers represented at the meeting indicated that they would be willing to provide trend information on a percentage basis but probably not raw sales data. It was agreed that confounding factors such as seasonality and promotions would need to be taken into account. Retailers agreed to explore how data could be 'smoothed' to take account of distortions from confounders like seasonality and promotions. Retailers agreed to suggest 'indicator products' that could be tracked to identify an appropriate timescale in order to record behavioural change.

Loyalty card data - it was felt that the significant practical issues and cost of data analysis would make the use of loyalty card data prohibitive. Till receipt data could provide a better source of information and warranted further exploration. Karen Tonks agreed to provide the group with a copy of a research report that used till receipt data.
Action: Karen Tonks

Focus groups - it was felt that these provided useful information on opinions and beliefs but were of limited value as a mechanism for exploring actual behaviour. Alternative methods of qualitative research such as food diaries and in-depth interviews were considered more appropriate. It was agreed that some sort of qualitative assessment could provide contextual insight that would help improve understanding of consumer behaviour, (See below).

Questionnaires - were felt to have limited value since reported behaviour often was very different to actual behaviour. If they were to be used then the questions would need to be carefully constructed and well focused.

Accompanied shops and cupboard audits - were felt to be useful as they could provide a deeper understanding of whether signposting was working or not.

Other ideas - it was agreed that the TNS Family Food Panel could potentially add helpful data. Observations and intercept interviewing were also considered useful approaches, although care would need to be taken to minimise the risk of bias resulting from the type of person likely to volunteer to take part.

Two potential options for management of the project were discussed. These were:

• An independently chaired panel with majority independent membership oversees the process. FSA, representative of each scheme being researched and consumer and health NGO representation also on the panel.
• Independent panel oversees the process. Observer status for FSA, each scheme being researched and consumer and health NGOs.

There was support for each option although it was generally felt that a list of independent members and possible chairpersons would be needed before making a final decision on the approach that should be taken. It was agreed that:

• FSA would approach independent researchers invited to the meeting, SACN and ACR, for suggestions for panel / chair positions to manage the project
• FSA would draw up and circulate list for comment, those present would be asked to indicate their preferred governance option for oversight of project
• experts to be approached for panel membership would need to be unanimously agreed
• members needed to have relevant research experience

Action: FSA to draw up and circulate list for comment, members also to be asked to indicate their preferred governance option for oversight of project.

The group considered how the term 'independent' should be defined. The following suggestions were made:

• Not a direct employee of any company or organisation with a direct interest in the results of the project. Potential persons being considered for a role in managing the project would be asked to declare any interests before being accepted as members
• Person should not have formally declared their position on signposting
• Should not be being paid directly by any companies or NGOs with a direct interest for the past or next 12 months

There was some general discussion on the timescale that the project would deliver over. Rosemary estimated that we might be able to start seeing some results within 12 to 18 months of the project starting but that full results might take longer. It was suggested that a comparison between products sold by retailers that use signposting and those that don't might compress the timescale and provide useful baseline data.

Two options for funding were identified:

• joint funding between sponsors of FOP signposting schemes and the FSA
• funding by FSA only

There was unanimous support for the FSA to fund the work, although retailers felt they could contribute to the project in kind by providing sales data. There was some suggestion that this could be open for reconsideration if it could be proved that industry funding would substantially improve the project's ability to deliver. All parties felt that for the study to be widely accepted it was necessary for it to be seen to be independent and independently funded. There was a desire to avoid consumer perception that industry funding was influencing the results.

It was agreed that those managing the project would need to keep aware of other activity that was going on in this area, including Department of Health work, the NCC report in July and back of pack initiatives. There was some discussion about whether the study should include the BCCCA 'treatwise' scheme. It was generally concluded that 'treatwise' did not fit very comfortably into the project's remit of looking at front of pack simplified nutritional information. While some group members thought it was important for the project to assess schemes that direct consumers to back of pack information, others were concerned that if the remit of the project was extended it could become unwieldy. It was agreed that the request would be noted in the minutes and that further consideration would be given to how front of pack logos and back of pack schemes might be tested separately.

FSA agreed to circulate a list of suggested candidates for the project management panel for agreement and seek formal views on the governance options at the same time.
Action: FSA