An experimental food label applied to food served at a secondary school in Greater London and its influence on adolescent food choices

Tucker, Joanne, Tsiami, Amalia ORCID: and Stock, Rosemary (2022) An experimental food label applied to food served at a secondary school in Greater London and its influence on adolescent food choices. In: 12th International Conference on Culinary Arts and Sciences 2022, 1-3 Jun 2022, Lyon, France.

Full text not available from this repository. (Request a copy)


Introduction: Adolescent obesity is rising in the United Kingdom (UK), and it is more prevalent in older adolescents aged 15-19 years than in almost any other higher income nation. Obese children and adolescents are at a high risk of developing several chronic diseases (NHS Digital, 2021; Nuffield trust, 2021). Providing nutritional information (NI) in the form of a food label (FL) is one strategy that has the potential to reduce obesity. A FL allows consumers to compare the NI between similar food products which can theoretically guide consumers as to which are the healthier food choices. Since December 2016 all prepacked food in the UK was required to display mandatory NI on the back of a FL, presented per 100g/ml or per portion of the product. The NI includes energy in kJ, and kcal and the number of grams of fat, saturated fat, carbohydrates, protein, and salt. NI on the front of prepacked FL is voluntary and is predominantly assessed through a nutrient profiling model (NPM) which can be a nutrient specific or a summary system. The process of NP includes several stages dependent upon its purpose. To date, there are a plethora of different front of pack FL designs yet no consensus as to which design has the greatest influence on food choice behaviour (FCB).

School food can theoretically provide a third of an adolescent daily calorie intake, yet schools in England have no policies to inform pupils as to which food options are the healthiest. The UK government has announced that from April 2022 it will enforce all businesses with 250 or more employees in England including cafes, restaurants and take away food to display calorific information on non-prepacked food and soft drinks (GOV.UK, 2021) but this does not include school food. There are three known published studies (Hunsberger et al., 2015; Rainville et al., 2010; Conklin and Cranage, 2005), and one unpublished study (Fresques’, 2013) that have provided NI in the form of calories and fat in a live school setting in America. Results did not concur as to the effectiveness of providing this NI and it can be suggested that due to possible adolescent vulnerability, a more overarching holistic FL is needed.

Objective: The objective of this study is to design and apply an experimental food label (EFL) to the food on offer in a secondary school setting in Greater London and to explore if it influences adolescents to select the healthier food options. Using a questionnaire this study also intended to identify if adolescents consider other factors are more important in their FCB and to evaluate if these factors require greater attention and integration at policy level to reduce adolescent obesity.

Method and Design: The study used a quantitative experimental design in which an intervention was implemented to evaluate if a change in FCB occurred. Several of the poorest borough councils in London were contacted to obtain a list of caterers who provide their school meals. One contract caterer agreed to take part in the study and nominated one school in Greater London in which to apply the EFL. The caterer provided the school’s three-week menu cycle, and their dietician provided a standardised recipe for each item on the menu. The information from the standard recipes was entered into the Diet plan 7 nutrition software package that created a breakdown of nutrients per 100g.

The Food Standards Agency developed a nutrient profiling model (NPM) for the Office of Communication (Ofcom). The aim of Ofcom`s NPM was to score food products, if foods scored 4 or more points, and drinks scored 1 or more points, they were classified as ‘less healthy’ and these products were not allowed to be advertised on television. This research used the Ofcom NPM and modified it to score food from one to 100 with a score of one being the least healthy and a score of 100 the healthiest. When the breakdown of nutrients per 100g from Diet plan 7 were input into the modified NPM this generated a food score for each dish on the menu cycle. The food score was the basis of the EFL.

The caterer recorded daily sales for each dish on the menu cycle for the nine-week intervention. Weeks 1, 2 and 3 were recorded as pre-intervention baseline sales. No scores were visible to the pupils at this stage. The week (0) before the intervention began tutors received an `awareness pack` which contained a standardised tutor sheet to read to their tutor group on how the food score worked and each pupil had a `know your score` postcard. The message on the post card was `the higher the score the healthier the food choices`. When the intervention commenced the food scores were placed in holders on the sneeze screen above each dish and A3 posters, identical to the pupils’ postcards, were displayed in the school restaurant. Pre-intervention baseline sales were compared to a repeat of the three-week menu cycle at post-intervention phase one (weeks 4, 5 and 6) and post-intervention phase two (weeks 7, 8 and 9). Sales were analysed to establish if the intervention influenced food choice behaviour and to evaluate if any effect would be sustained. It was expected that the mean scores of the foods selected will increase after the application of the food score, that will indicate that pupils selected healthier food choices.

After the nine-week intervention pupils were requested to complete a four-part paper-based questionnaire. Part one requested sociodemographic data, part two asked lunch preferences, part three asked participants to rank on the five-point scale whether they noticed the food score, and their understanding of the food score. Part four asked participants to rank (using the same five-point scale) which factors from the combined food choice questionnaire (FCQ) (Steptoe et al., 1995 and Ooi et al., 2015) they felt has the greatest impact on their FCB at school.

Results: Each dish on the menu cycle had a calculated food score. Each time a pupil selected a dish the food scores were recorded to provide a daily score for each dish. The daily scores were added together to provide a mean food score for each dish each week. If after the application of the EFL healthier food options (higher scores) were selected more frequently the mean food scores would increase. The differences in the mean food score before and after the application of the EFL would indicate its effectiveness.

From the nine week experiment the mean food score from pre-intervention baseline (weeks 1, 2 and 3) (43587.21) to post-intervention phase one (week 4, 5 and 6) (47473.93) increased by 3886.72 then decreased from post-intervention phase one (47473.93) to post-intervention phase two (weeks 7, 8 and 9) (46454.64) by 1019.29. The total increase in the food score from pre-intervention baseline to post-intervention phase two was 2867.43. Mauchly`s test indicated significance at 0.002, so the Greenhouse Geiser adjustment was used, and an ANOVA indicated that, although the mean suggests an upward trend, no significant difference was present (F(2, 26) = 2.19, p > .05).

Literature suggests that females are more interested in health and reading FL more than males. This research sought to detect if females reported noticing the food score more than males and if females self-reported that the food score affected their food choices more than males. In the FCQ pupils were asked to select (from a five-point scale) how much they noticed the food score from one, they did not notice to five, they did notice. Using the same scale, pupils were asked how much the food score affected their food choices from one, it did not affect my food choice to five, it did affect my food choices. From the 744 pupils’ response to `Did you noticed the food score` then removing the incomplete data for gender declaration, 674 responses remained. From the 346 female responses, 198 (57.2%) selected score 1 meaning they did not notice the food score and 54 (15.6%) selected score 5 meaning that they did notice the food score. From the 328 male responses, 186 (56.7%) selected score 1 and 61 (18.6%) selected score 5. There was a non-significant association between participant gender and noticing the food score χ² (4) = 1.46, p > .05. From the self-reported questionnaire fewer pupils responded to the question of whether the food score affected their food choices. From the 661 pupils’ response to `Did the food score affect your food choices` From the 339 female responses, 216 (63.7%) selected score 1 meaning it did not affect their food choices and 25 (7.4%) selected score 5 meaning that it did affect their food choices. From the 322 male responses, 216 (67.1%) selected score 1 and 19 (5.9%) selected score 5. There was a non-significant association between pupils’ gender and pupils reporting that the food score had a perceived effect on their food choices χ² (4) = 1.36, p > .05 which does not support the current literature.

The research aimed to ascertain which of the 21 factors on the FCQ adolescents score as the most (score five) and the least important (score one) regarding influencing their FCB at school. Results indicated that the highest percentage of pupils scored the following factors at score five, 64.9% `tastes good`, 32.9% `keeps me full`, 29.9% is cheap, 29.2% keeps me healthy, 28.9% looks nice, 26% smells nice, and 25.2% keeps me awake. The factor that was identified as the least influential (score one) was 66.5% `is similar to foods chosen by my friends`.

The research sought to identify if there was an association between gender in food choice and the importance placed on health. Literature advocates that females are more influenced by health than males. Using the FCQ with the five-point scale `health and nutrition` encompassed nine factors; contains vitamins and minerals, is nutritious, is high in fibre, is high in protein, is low in calories, is low in fat, helps me control my weight, keeps me healthy, and includes lots of fruit and vegetables. Results indicated that there was a significant association between gender and specific health and nutrition factors. From 689 participants (361 female and 328 male) females rate `contains vitamins and minerals` significantly more highly than males χ2(4) = 10.4, p < .05. Females rate `is nutritious` significantly more highly than males χ2(4) = 11. 17, p <. 05. Females rate the factor `is low in calories` significantly more highly than males χ2(4) = 9.62, p < .05. Females rate `is low in fat` significantly more highly than males χ2(4) = 11.09, p < .05. Females rate `keeps me healthy` significantly more highly than males χ2(4) = 21.38, p < .05. It could be suggested that the word `health` is used to promote or identify the healthier food choices which may appeal to females.

Conclusion: A food score applied to food served in a secondary school in Greater London was not effective in influencing FCB. Only 15.6% females and 18.6% males noticed the food score and 7.4% females and 5.9% males stated that it influenced their FCB. There were no differences in gender as to noticing or self-reporting any effect on their FCB. Adolescents rated sensory aspects of taste, looks, and smell alongside satiation, keeping awake and financial aspects at score of five. Food choice and the importance placed on health was significantly associated more with females than males. The influence of friends was not considered to be a significant factor in adolescent FCB.

Item Type: Conference or Workshop Item (Lecture)
Subjects: Hospitality and tourism > Culinary arts
Hospitality and tourism > Culinary arts > Food service operations
Hospitality and tourism > Culinary arts > Food studies
Related URLs:
Depositing User: Amalia Tsiami
Date Deposited: 23 May 2022 17:25
Last Modified: 06 Jun 2022 09:57

Actions (login required)

View Item View Item