Understanding Food Choices in Bahrain Using Bahrain e-Mart

Experimental design \& procedures

The aim of this study is to use a 2-arm randomized controlled trial with an online grocery store (Bahrain e-Mart), to rigorously evaluate the effect of the Multiple Traffic Light front-of-pack (MTL FOP) labels displayed on all food and beverage products, on diet quality.

Multiple Traffic Light (MTL) Labels

The investigators will test one type of front-of-pack (FOP) label known as MTL. The MTL label includes per serving size information and grades each nutrient i.e., energy, sugar, fat, saturated fat, and sodium separately based on recommended thresholds. Green signifies a healthy amount of that nutrient; red signifies an unhealthy amount, and amber signifies that the nutrient levels fall between healthy and unhealthy amounts. Additionally, the label also shows how much of a person's daily allowance for a particular nutrient is met by consuming one serving of the product. Lastly, MTL includes the absolute values of each nutrient per serving of a product and the percentage of an adult's daily reference intake that is met by consuming a serving of this product.

Overview of Randomised Control Trial (RCT) Design

To test these labels, the investigators used two different versions of Bahrain e-Mart. Each participant was randomly assigned to one of the following arms and asked to complete a one-time shop.

Arm 1 (Control): Participants will experience a default version of Bahrain e-Mart which replicates the traditional shopping experience of online grocery stores with no FOP labels.

Arm 2 (MTL): Same as Arm 1 Bahrain e-Mart except that Multiple-Traffic Light (MTL) labels are displayed on food and beverage products. Multiple Traffic Light is a nutrition labelling system wherein each nutrient attribute constituting this label (i.e. sugar, saturated fat, fat, and sodium) is assigned different colours according to whether the amount of that nutrient is low (green), medium (amber) or high (red).

To collect shopping data as close as possible to shoppers' actual grocery carts, the investigators set a minimum spending value per person per week and adjusted the total minimum spending value depending on household size. Additionally, participants will have to shop from at least 4 different Bahrain e-Mart categories to successfully checkout their cart. Finally, participants will be informed that they may win rebates ranging from 25% to 100% on their grocery order. This will be done by implementing an electronic prize wheel that the participants will spin after successfully checking out their cart. Every participant will have an equal chance of winning. If they win any of the rebates, they will be expected to conduct a grocery shop in a market/supermarket of their choice to purchase the same items ordered on the Bahrain e-Mart grocery store website. The maximum rebate they can get after spinning the wheel will be based on their drawn rebate rate and the total order amount on Bahrain e-Mart (e.g., If participants win a 50% rebate and the total order amount on Bahrain e-Mart was BD 40, the maximum rebate they can get would be BD 20). If the exact same item cannot be found, they are allowed to purchase a similar item as a replacement instead. Replacements are subject to the guidelines provided in the participant information sheet and consent form that they must sign prior to enrolment and randomisation into either of the study arms.

Subject related procedures

Participants will be recruited if they are Kingdom of Bahrain residents aged 21 years or older, can speak and write Arabic or English and are primary weekly shoppers for their households. Recruitment will be done by a market research company utilising in-person intercept surveys. This survey aims to recruit participants across 4 cities and 20 locations (e.g., shopping malls). The Duke-NUS team is not directly involved in data collection.

Participants will be intercepted in-person by the interviewer and be briefed about the study. Those who are interested will first be invited to select their preferred language (i.e. English or Arabic) and then complete the online screener questionnaire using a tablet. All eligible participants will then be asked to enter their mobile number in a textbox field and will be required to give consent for the investigators to use their personal data for registration purposes, that is, to verify via OTP that their mobile number has never been entered into the system before. This ensures that the participant is not a duplicate participant and has not attempted to join the study before. Participants who decide to withdraw from the study after screener completion and provision of their mobile number will be ineligible to participate in the study again.

Participants will subsequently be asked to read an information sheet and provide their consent to enroll in this study by entering their name and email address in a textbox field. Upon consent, participants will be redirected to complete a baseline questionnaire to collect demographic and health characteristics, which should take approximately 10 minutes to complete. The baseline questionnaire includes a question as to whether any household members have a medical condition, such as diabetes or hypertension, which requires limiting the types of foods they eat. Investigators ask this question to allow for testing whether the intervention differentially influence these households, with the expectation that households with less healthy patients may obtain greater benefits from purchasing baskets with higher mean Nutri-Scores. Since the objective of this (RCT is to quantify the effectiveness of the intervention on diet quality, a precursor to non-communicable diseases (NCDs), the collection of household health indicators is reasonable.

Upon completion of the baseline survey, participants will be randomly assigned to one of the 2 shopping conditions (Arm 1 [Control] or Arm 2) and redirected to Bahrain e-Mart to log in and begin shopping immediately.

Participants will be informed that the shop must be completed with the aim of purchasing enough groceries for all members of their household for a week. If they have a household size of 8 members or less, they are required to spend a minimum of BD 10 (≈USD 26) for each member of their household, and their expenditure should not exceed twice the total minimum. For instance, if they have 4 members their minimum expenditure should be BD 10 X 4 = BD 40 and their maximum expenditure should be BD 40 x 2 = BD 80. However, if they have a household size of more than 8 members, they are required to spend a minimum of BD 80 and a maximum of BD 160. Additionally, participants must select products from at least 4 different Bahrain e-Mart Store categories (e.g. Dairy and eggs, fruits and vegetables etc.).

The interviewer will stand at a distance to allow participants to complete their shop privately up until the prize wheel, unless the participant raises any questions. The time spent completing the shop is expected to be approximately 15-20 minutes. Subsequently, participants will be required to complete a post-study survey which should take around 3-5 minutes to complete.

Upon completion of the post-study survey, participants will spin a random electronic prize wheel. Depending on the results of the prize wheel, their shopping trip may or may not involve actual purchases. Participants may win rebates ranging from 25% to 100% based on their total order amount on Bahrain e-Mart.

To claim their rebate, participants are expected to email the market research company ONE image of their purchased grocery items and ONE clear image of the itemized grocery receipt for verification within 14 days from the time of study completion, which is after the participant spins the prize wheel. If no claims have been received via email within the stipulated time, the rebate will be forfeited. Please note that participants will still receive BD 2.5 via online transfer as stated in the participant information sheet.

Once the market research company validates the two images above, the study reimbursement and rebate amount will be transferred to the participant via online transfer. This will be done within 24 working hours of the company receiving the images.

If the prize wheel lands on "BD 2.5", then no further action is required. The study reimbursement will be transferred to the participant via online transfer immediately upon study completion. The study reimbursement of BD 2.5 will be given to all participants, regardless of the outcome of the prize wheel. The prize wheel only determines the additional rebates that participants may win.

All individuals who provided their consent to join this study will be emailed a copy of the full information sheet and consent form as well as the debriefing sheet when they complete the study (i.e., after participant has spun the wheel of rebate) or if they choose to withdraw from the study. This is the same for participants who land on "BD2.5". The interviewer will remind participants to review the debriefing sheet on their own.

Analysis Plan

An ordinary least square model will be used to compare the primary and secondary outcome variables between the MTL arm and the control arms. The model will be adjusted for potential confounders including age, gender, household size, education level, income, and prevalence of diet-related health conditions that may affect the outcomes.