The impact of the UK soft drink industry levy on the soft drink marketplace, 2017-2020: An interrupted time series analysis with comparator series.

BACKGROUND: In April 2018, the UK government implemented a levy on soft drinks importers and manufacturers, tiered according to the amount of sugar in drinks. The stated aim was to encourage manufacturers to reduce sugar and portion sizes. Previous evidence suggests that the policy has been successful in reducing sugar in drinks in the short-term since implementation, but their sustained effects have not been explored. This study aimed to assess the impact of the soft drink industry levy (SDIL) on sugar levels, price, portion size and use of non-sugar sweeteners in the medium-term. METHODS AND FINDINGS: Product data from 30 November 2017 to 14 March 2020 from one major UK retail supermarket were analysed (112,452 observations, 126 weekly time points). We used interrupted time series analysis, to assess the impact of the soft-drink industry levy (SDIL) on levy-eligible soft drinks, with exempt drinks (i.e. 100% fruit juices, milks, flavoured milks) acting as a comparator series. At the point of implementation of the SDIL (April 2018) there was a step change in the proportion of eligible drinks with sugar content below the SDIL levy threshold (5g per 100ml) (+0.08, 95%CI: +0.04, +0.12), with a similar sized decrease in the proportion in the highest levy category (> = 8g sugar per 100ml) (-0.06, 95%CI: -0.10, -0.03). Between April 2018 and March 2020, the proportion of eligible drinks below the SDIL levy threshold continued to gradually increase (p = 0.003), while those in the highest levy category decreased (p = 0.007). There was a step change in price of eligible drinks in the higher levy category at the point of implementation of +£0.049 (95%CI: +£0.034, +£0.065) per 100mL (for comparison, the levy is set at £0.024 per 100mL for this group). Trends in price for the high levy category were not altered by the SDIL. In the no levy category, there was a step change in price at the implementation (+£0.012 per 100mL, 95%CI: +£0.008, +£0.023), followed by a second step change in October 2018 (-£0.018p per 100mL, 95%CI: -£0.033, -£0.001p). The volume of products in the higher levy group decreased at the time of the implementation (-305mL on average including multipacks, 95%CI: -511, -99). The change in trend for the product volume of drinks in the higher levy group between April 2018 and March 2020 was in the increasing direction (+704mL per year, 95%CI: -95, 1504), but it did not meet our threshold for statistical significance (p = 0.084). There were no changes observed in the volume of lower levy drinks or no levy drinks. There was a step change in the proportion of drinks with non-sugar sweeteners at the implementation of the SDIL (+0.04, 95%CI: +0.02, +0.06). CONCLUSION: These results suggest that the SDIL was successful in [1] producing reductions in sugar levels that were maintained over the medium term up to March 2020 and [2] a reduction in product volume for higher tier drinks that may be diminishing over time. Our results also show that the SDIL was associated with a maintained price differential between high and low sugar drinks.

Do promotions of healthier or more sustainable foods increase sales? Findings from three natural experiments in UK supermarkets.

BACKGROUND: Dietary changes are necessary to improve population health and meet environmental sustainability targets. Here we analyse the impact of promotional activities implemented in UK supermarkets on purchases of healthier and more sustainable foods. METHODS: Three natural experiments examined the impact of promotional activities on sales of a) no-added-sugar (NAS) plant-based milk (in 199 stores), b) products promoted during 'Veganuary' (in 96 stores), and c) seasonal fruit (in 100 non-randomised intervention and 100 matched control stores). Data were provided on store-level product sales, in units sold and monetary value (£), aggregated weekly. Predominant socioeconomic position (SEP) of the store population was provided by the retailer. Analyses used interrupted time series and multivariable hierarchical mixed-effects models. RESULTS: Sales of both promoted and total NAS plant-based milks increased significantly during the promotional period (Promoted:+126 units, 95%CI: 105-148; Overall:+307 units, 95%CI: 264-349). The increase was greater in stores with predominately low SEP shoppers. During Veganuary, sales increased significantly for plant-based foods on promotion (+60 units, 95%CI: 37-84), but not for sales of plant-based foods overall (dairy alternatives: -1131 units, 95%CI: -5821-3559; meat alternatives: 1403 units, 95%CI: -749-3554). There was no evidence of a change in weekly sales of promoted seasonal fruit products (assessed via ratio change in units sold: 0.01, 95%CI: 0.00-0.02), and overall fruit category sales slightly decreased in intervention stores relative to control (ratio change in units sold: -0.01, 95%CI: -0.01-0.00). CONCLUSION: During promotional campaigns there was evidence that sales of plant-based products increased, but not seasonal fruits. There was no evidence for any sustained change beyond the intervention period.

Changing the availability and positioning of more vs. less environmentally sustainable products: A randomised controlled trial in an online experimental supermarket.

Food purchasing behaviours are shaped by the choices available to shoppers and the way they are offered for sale. This study tested whether prominent positioning of more sustainable food items online and increasing their relative availability might reduce the environmental impact of foods selected in a 2x2 (availability x position) factorial randomised controlled trial where participants (n=1179) selected items in a shopping task in an experimental online supermarket. The availability intervention added lower-impact products to the regular range. The positioning intervention biased product order to give prominence to lower-impact products. The primary outcome was the environmental impact score (ranging from 1 "least impact" to 5 "most impact", of each item in shopping baskets) analysed using Welch's ANOVA. Secondary outcomes included interactions (analysed via linear regression) by gender, age group, education, income and meat consumption and we assessed intervention acceptability (using different frames) in a post-experiment questionnaire. Compared to control (mean=21.6), mean eco quintile score was significantly reduced when availability & order was altered (-2.30; 95%CI: -3.04; -1.56) and when order only was changed (-1.67; 95%CI: -2.42; -0.92). No significant difference between availability only (-0.02; 95%CI: -0.73; 0.69) and control was found. There were no significant interactions between interventions or by demographic characteristics. Both interventions were acceptable under certain frames (positioning emphasising lower-impact products: 70.3% support; increasing lower-impact items: 74.3% support). Prominent positioning of more sustainable products may be an effective strategy to encourage more sustainable food purchasing. Increasing availability of more sustainable products alone did not significantly alter the environment impact of products selected.

The development of a method for the global health community to assess the proportion of food and beverage companies' sales that are derived from unhealthy foods.

CONTEXT: Corporate engagement with food and beverage companies who produce food associated with health harms is a divisive topic in the global nutrition community, with high-profile cases of conflict of interest increasingly coming under scrutiny. There is a need for an agreed method to support health organizations in deciding whether and how to engage with large food and beverage manufacturers. AIM: The aim of this study was to develop a method to quantify the proportion of sales from food and beverage companies that are derived from unhealthy foods to support organizations in determining which companies might be considered high-risk for engagement. METHODS: The 2015 WHO Euro nutrient profile model was applied to 35,550 products from 1294 brands manufactured by the top 20 global food and beverage companies from seven countries (Australia, Brazil, China, India, South Africa, UK and USA). For the purpose of this study, products that met the WHO Euro criteria were classified as "healthier" and those that failed were classified as "unhealthy". Products were grouped by brand and weighted by the brand's value sales for 2020. The primary outcome was the proportion of each company's sales that were classified as unhealthy and healthier by company and category. RESULTS: Overall, 89% of the top 20 companies' brand sales were classified as unhealthy. For every USD$10 spent on the top 20 companies' brands, only $1.10 was spent on products considered healthier. All companies saw the majority of their sales come from unhealthy foods, including soft drinks, confectionery and snacks. None of Red Bull or Ferrero's sales were classified as healthier and less than 5% of total sales were healthier for Mondelez, Mars, and PepsiCo. Some companies had higher proportions of sales deriving from healthier products, including Grupo Bimbo (48%), Danone (34%) and Conagra (32%), although the majority of their sales were still derived from unhealthy foods. DISCUSSION: The results presented in this study highlight the reliance the leading food and beverage companies have on sales of unhealthy products that are contributing to diet-related disease globally. The method and steps we have laid out here could be used by organizations in the global health community to identify companies that have conflicts of interest when it comes to engaging with governments, international organizations and public health bodies on issues of policy and regulation.

The impact of altering restaurant and menu option position on food selected from an experimental food delivery platform: a randomised controlled trial.

BACKGROUND: Overconsumption is one of the most serious public health challenges in the UK and has been linked to increased consumption of food ordered through delivery platforms. This study tested whether repositioning foods and/or restaurant options in a simulated food delivery platform could help to reduce the energy content of users' shopping basket. METHODS: UK adult food delivery platform users (N = 9,003) selected a meal in a simulated platform. Participants were randomly allocated to a control condition (choices listed randomly) or to one of four intervention groups, (1) food options listed in ascending order of energy content, (2) restaurant options listed in ascending order of average energy content per main meal, (3) interventions 1 and 2 combined (4) interventions 1 and 2 combined, but food and restaurant options repositioned based on a kcal/price index to display options lower in energy but higher in price at the top. Gamma regressions assessed the impact of interventions on total energy content of baskets at checkout. RESULTS: The energy content of participants' baskets in the control condition was 1382 kcals. All interventions significantly reduced energy content of baskets: Compared to control, repositioning both foods and restaurants purely based on energy content of options resulted in the greatest effect (-209kcal; 95%CIs: -248,-168), followed by repositioning restaurants (-161kcal; 95%CIs: -201,-121), repositioning restaurants and foods based on a kcal/price index (-117kcals; 95%CI: -158,-74) and repositioning foods based on energy content (-88kcals; 95%CI: -130,-45). All interventions reduced the basket price compared to the control, except for the intervention repositioning restaurants and foods based on a kcal/price index, which increased the basket price. CONCLUSIONS: This proof-of-concept study suggests repositioning lower-energy options more prominently may encourage lower energy food choices in online delivery platforms and can be implemented in a sustainable business model.

The impact of price promotions on confectionery and snacks on the energy content of shopping baskets: A randomised controlled trial in an experimental online supermarket.

Overconsumption of foods high in fat, sugars, and salt (HFSS) poses a significant risk to health. The government in England has passed legislation that would limit some price promotions of HFSS foods within supermarkets, but evidence regarding likely impacts of these policies, especially in online settings, is limited. This study aimed to determine whether there were any differences in the energy and nutrient content of shopping baskets after removing promotions on HFSS foods in an online experimental supermarket. UK adults (n = 511) were asked to select food from four categories with a £10 budget in an online experimental supermarket: confectionery; biscuits and crackers; crisps, nuts and snacking fruit; cakes and tarts. They were randomly allocated to one of two trial arms: (1) promotions present (matched to promotion frequency seen in a major UK retailer) (n = 257), or (2) all promotions removed from all products within the target food categories (n = 254). The primary outcome analysis used linear regression to compare total energy (kcal) of items placed in shopping baskets when promotions were present vs. absent, while secondary analyses investigated differences in nutrients and energy purchased from individual food categories. Mean energy in food selected without promotions was 5156 kcal per basket (SD 1620), compared to 5536 kcal (SD 1819) with promotions, a difference of -552kcal (95%CIs: -866, -238), equivalent to 10%. There were no significant differences in energy purchased for any individual category between groups. No evidence was found of other changes in nutritional composition of baskets or of significant interactions between the impact of promotions and participant characteristics (gender, age, ethnicity) on energy purchased. Removing promotions on HFSS foods resulted in significantly less total energy selected in an online experimental supermarket study.