Supermarket Labelling and Taxation Study

The study will be a 2x2 (labelling and taxation) between-subjects randomised controlled trial in which participants will be recruited assigned to one of four conditions in the online survey platform "Qualtrics": (i) nutrient warning labelling and no taxation, (ii) nutrient warning labelling and taxation, (iii) current traffic light labelling and taxation or (iv) a control condition (current traffic light labelling and no taxation). In the nutrient warning labelling and no taxation condition, the corresponding warning labels will be placed on items that are high in saturated fat, salt, sugar, and/or calories. Items will be defined as 'high in' if they exceed >25% of the reference intake (RI) /100g or >30% RI/portion, consistent with the threshold for a red traffic light. RIs are established in European law (EU Regulation 1169/2011) and reflect the daily requirements of an average adult. In addition to these criteria, items will be defined as 'high in' calories if they exceed Public Health England (now UKHSA) guidelines for a meal (>600kcals/portion) or snack (>200kcals/portion). In the nutrient warning label and taxation condition, a 20% price increase (based on modelling work indicating efficacy at this level for reducing energy intake and obesity prevalence in a UK context) will also be applied to 'high in' items. In the traffic light and no taxation condition, all items will feature a traffic light label. This is broadly in line with current UK implementation (an estimated 75% of packaged products feature the label under voluntary implementation). In the traffic light label and taxation condition, a 20% price increase will also be applied to 'high in' items.

Participants will complete a food ordering task using a simulated online supermarket resembling a major UK retailer (Tesco). A subsample of items from three categories - Ready Meals for One, Starters and Sides, and Chilled Desserts - will be available for selection. All items will be from the "Tesco" brand. The subsample (8 items per category) was selected to be representative of the full sample in terms of (i) item range (i.e., cuisine type, including complementary combinations) and (ii) healthiness (i.e., similar proportions of items that are low, medium, and high across the nutrients of concern, and high/not high in energy). Ready meals that contained >1 portion, starters/sides that contained >2 portions, and desserts that contained more than >1 portion and portions that were not clearly designated (e.g., individual slices, packs) were excluded to reduce variability in item size within each category, as this could add noise to intake outcomes. The name, price, and image of the item will be displayed, consistent with the Tesco website. Item images can be clicked to display further information, such as the product description, ingredients, and nutritional information (including enlarged label), aligned with information that is presented on the Tesco website.

Participants will either be recruited via the online participant panel platform "Prolific" or via social media/a database maintained by researchers at the University of Liverpool. Participants will answer brief demographic questions, including age, sex, gender, highest education level, and annual post-tax household income (open response). Any participants completing the survey outside of Prolific will be asked to create a personal ID (first name initial, last name initial, last two digits of their phone number) which they will enter in both this survey and the follow-up survey, so that their responses can be linked. For Prolific participants, they will enter their unique Prolific ID. An attention check prompting participants to select a specific answer will be included here.

After answering demographic questions, participants will be shown instructions for the food ordering task: On the next page, a Tesco-style food selection interface will be displayed. Please choose the item(s) for yourself to be consumed for an evening meal within the next week. You are required to select one main, and can also optionally select a side, and/or a dessert. Bear in mind that you will be required to actually order, purchase, and consume these item(s).

Once participants have read the instructions, they will proceed to the Tesco-style food selection interface. Participants will be required to select one main, and, optionally, a side and/or dessert, indicated again by an on-screen "shopping list" that ticks off items as they are selected. Participants can add/remove items, navigate back-and-forth between main, side, and dessert options when making their selections, and see their selections populate a "basket" before checking out, in an interface that has been built to closely resemble the Tesco website.

After checking out, a basket summary will be displayed to participants indicating their selections. They will then be asked to open a new tab (or click the "open Tesco in a new tab" button in the survey, which opens Tesco in a new tab), and order their selected items on the Tesco website for collection/delivery within the next week. They will be told that if any of their items are unavailable, they should select what they deem to be the next-closest item available. Note - prices displayed on the Tesco website will not match those in the taxation conditions. However, we do not anticipate this to be an issue, as they will always be paying the same/less than indicated in their basket summary, and it is their initial choice that we are primarily interested in.

Participants will then be requested to upload an image/screenshot of their receipt/order confirmation, clearly indicating the items and delivery time and date. They will be asked to not include personal information (e.g., their name) in the upload. If any uploads contain personal information, this will be redacted by the researcher. They will also be asked to confirm (i) when they will receive their food (date and time) and (ii) when they will eat their food (i.e., evening of the [date]), specifying that they must adhere to this date, to aid scheduling of the follow-up survey.

After completing the ordering task, participants will answer further questions on aim guessing (open response), label and price awareness, food choice motives, representativeness of their order and the interface, and policy support. Any participants completing the survey outside of Prolific will be asked to provide their email address so that they can be contacted for the follow-up survey. Before submitting the survey, the following instructions will be displayed to participants: Thank you for answering the questions and placing your Tesco order. You will be invited to participate in a follow-up survey via [Prolific/email] the morning after you have eaten the food item(s) for your evening meal. You will receive [a reimbursement] once you have completed the follow-up survey. Please take photos of your food items before (i.e., when they are plated) and after you have finished eating, as we will ask you to upload these images in the follow-up survey. You can click the "download as PDF" button to save these instructions for your reference. Don't forget to also click the arrow to submit your survey response. If you have any questions, please contact us via [Prolific/email]. The survey will take approximately 20 minutes to complete.

The morning after participants have eaten the food items for their evening meal, they will be invited via Prolific to complete a follow-up Qualtrics survey. They will be required to upload their meal images and complete dummy questions rating each item's flavour, texture, and value for money. They will also be requested to complete a dietary recall survey for breakfast and lunch the previous day via Intake24. A unique Intake24 link will be displayed to each participant via Qualtrics based on their Prolific/personal ID number, as a clickable "open new tab" button. Participants will then return to Qualtrics to confirm their Intake24 submission Upon completion, a debrief form will automatically be displayed to participants, detailing the true study aims, sources of support, and researcher contact details. The follow-up survey will take approximately 10 minutes to complete. Any participants completing the study outside of Prolific will be asked to confirm their email address in the follow-up survey to facilitate reimbursement.