Consumption of sugar-sweetened beverages (SSBs) like sodas, sports drinks, and fruit drinks remains a pressing public health concern in the United States. Consumption of SSBs remains well above recommended levels, and has been linked to obesity, diabetes, and cardiovascular disease. In response, policymakers across the U.S. have proposed requiring warning messages on SSB containers to inform the public and reduce consumption to healthier levels. While online studies find that such warnings reduce intentions to purchase SSBs, no studies have yet examined the impact of warnings on actual consumer behavior. The purpose of this randomized trial is to determine whether health warnings on sugar-sweetened beverages are more effective than control labels at reducing purchases of sugar-sweetened beverages. The trial will take place in a mock convenience store. Participants will be randomly assigned to a health warnings arm in which all sugar-sweetened beverages in the store are labeled with a health warning, or to a control arm in which all sugar-sweetened beverages in the store are labeled with a neutral label. Participants will select items to purchase with cash, and purchases will be compared across trial arms.
Background: Obesity, diabetes and cardiovascular disease are among the leading causes of death in the United States. Consumption of sugar-sweetened beverages (SSBs) such as sodas, fruit drinks and sports drinks increases risk of these preventable conditions. SSBs are particularly problematic because they are high in calories but offer little nutritional value, and because the liquid calories in SSBs cause faster spikes in blood sugar and lower feelings of fullness than solid foods. Despite these risks, the average American adult consumes more than 130 calories per day from SSBs, well above recommended levels. A promising strategy for reducing SSB consumption is requiring health warnings on SSB containers. Online studies show that health warnings attract consumers' attention and reduce intentions to purchase SSBs. However, no research has tested the impact of SSB health warnings on actual behaviors. This randomized trial evaluates whether health warnings reduce purchases of SSBs. Setting: The trial will take place a mock convenience store setting created for researchers to examine how store and product characteristics influence consumer purchases in a controlled but realistic environment. The store reflects a convenience store's characteristics, selling foods, beverages, and household items at real-world prices. Participants can buy these products using cash. Recruitment: Participants will be recruited via university community subject pools, printed flyers posted, and online listings. Interested potential participants will complete an online pre-enrollment screening questionnaire to determine their eligibility. If eligible, they will be invited to schedule a time to visit the mock store for the trial. Informed consent: At the beginning of the study visit, research staff will review the consent documents with the participants. After reviewing the form, the research staff member will ask the participant if he or she has any questions. Then both parties will sign the consent form and the participant will receive a copy of the consent form to keep. Randomization: Participants will be randomly assigned to trial arms. An independent biostatistician will conduct the randomization a priori. Participants will have an equal chance of being randomized to either arm of the trial. Randomization will be carried out by generating a random number for each participant identification code (ID). Participant IDs will be ordered on this random number from smallest to largest. The bottom half of the IDs will be assigned to the control arm and the top half to the treatment arm (trial arms are described below). When participants arrive for their study visit, they will be assigned the next consecutive participant ID and will be allocated to the trial arm previously assigned to that ID. Experimental Procedures: When participants arrive at the mock store, they will be provided with a shopping basket and their participation incentive in cash. Participants will be instructed to pretend they are on a usual shopping trip and reminded that there are no right or wrong products to select. They will not be told the trial's purpose. Participants will be asked to complete a shopping task. They will be given a list of product categories and asked to select their preferred products in each of these categories. They will indicate their selections by placing products in their shopping basket. Before they begin the shopping task, participants will be instructed that they will be required to purchase one of the items they select, and that this item will be chosen for them at random by the experimenter. This procedure incentivizes participants to only select products they are actually interested in purchasing. After participants have completed the shopping task, the experimenter will randomly choose one of the products in the participant's basket for participant to purchase with his or her incentive cash. The participant will pay for the product and will receive change as necessary. After paying for their item, participants will complete a computer-based survey in another room. Assessment: The products participants select will be recorded by the experimenter at the end of the shopping task. Participants will also complete 1 computer-based survey immediately after completing the shopping task. Detailed description of the trial arms: Participants will be assigned to either a health warnings arm or a control arm. In the health warnings arm, all SSB products will be labeled with a health warning label. Investigators developed the text and design of these labels based on previous research and on label designs described in US and international legislation. In the control arm, all SSB products will be labeled with a neutral label, also developed by the investigators. In both arms, labels will be adhered directly on the front-of-package of SSB containers.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
PREVENTION
Masking
SINGLE
Enrollment
400
Labels with a health warning will be applied to the front-of-package of all sugar-sweetened beverage containers in the mock store. Investigators developed the text and design of these labels based on previous research and on designs described in US and international legislation.
Neutral labels will be applied to the front-of-package of all sugar-sweetened beverage containers in the mock store. Investigators developed the design of these labels.
University of North Carolina Chapel Hill
Chapel Hill, North Carolina, United States
Mean Number of Calories (kcal) Purchased From Sugar-sweetened Beverages
The primary outcome is calories of sugar-sweetened beverages purchased during the shopping task. Calories from sugar-sweetened beverage purchases is defined as the total number of calories from sugar-sweetened beverages in the participant's basket when they complete the shopping task, calculated as the sum of calories/container for all SSB containers. Purchases will be recorded by the experimenter immediately after the participant completes the \~10-minute shopping task.
Time frame: At completion of ~10-minute shopping task
Mean Number of Calories (kcal) Purchased From All Foods and Beverages
Total calories purchased from all foods and beverages is defined as the total number of calories from all products (including sugar-sweetened beverages, non-sugar-sweetened beverages, and all foods) in the participant's basket when they complete the shopping task. Total calories purchased will be calculated as the sum of calories/container for all products. Purchases will be recorded by the experimenter immediately after the participant completes the \~10 minute shopping task.
Time frame: At completion of ~10-minute shopping task
Percent of Participants Who Purchase a Sugar-sweetened Beverage
Purchase of a sugar-sweetened beverage is defined as the participant having one or more sugar-sweetened beverage in their basket when they complete the shopping task. Purchases will be recorded by the experimenter immediately after the participant completes the \~10-minute shopping task.
Time frame: At completion of ~10-minute shopping task
Mean Number of Sugar-sweetened Beverages Purchased
Number of sugar-sweetened beverages purchased is defined as the number of sugar-sweetened beverages in the participants' basket when they complete the shopping task. Purchases will be recorded by the experimenter immediately after the participant completes the \~10-minute shopping task.
Time frame: At completion of ~10 minute shopping task
Mean Intentions Score to Limit Consumption of Beverages Added Sugar
Intentions to limit consumption of beverages with added sugar will be measured at post-test using 3 items adapted from Klein, Zajax \& Monin (2009). The items ask participants to rate the extent to which they want to, plan to, and are likely to, drink less than 1 beverage with added sugar in the next week. Responses to these three items will be averaged to create a mean intentions score. Response options are on a 1 to 7 scale, thus, mean scores can range from 1-7, with higher scores indicating higher intentions to limit consumption of beverages with added sugar.
Time frame: Within 30 minutes following completion of ~10-minute shopping task
Mean Intentions Sore to Limit Consumption of Specific Sugar-sweetened Beverages
Intentions to limit consumption of specific beverage categories (i.e., soda, fruit drinks, sport drinks/flavored waters, sweetened coffee/tea, and energy drinks) will be measured at post-test using measures adapted from Klein, Zajax, \& Monin (2009). For each of the five beverage categories, participants will rate the extent to which they are likely to drink less than 1 serving of the beverage in the next week. Responses will be averaged to a create a mean intentions score. Response options are on a 1 to 7 scale. Thus, mean scores can range from 1-7, with higher scores indicating higher intentions to limit consumption of these beverages.
Time frame: Within 30 minutes following completion of ~10-minute shopping task
Percent of Participants Who Notice the Trial Labels
Noticing of the trial labels will be measured at post-test using 1 item adapted from Roberto et al. (2016). The item asks participants to indicate whether they noticed the trial labels. Response options are 0 (no) and 1 (yes).
Time frame: Within 30 minutes following completion of ~10-minute shopping task
Mean Attention Score to the Trial Labels
Attention to the trial labels will be measured at post-test using 2 items adapted from Nonnemaker et al. (2010). The items ask participants to rate their agreement with statements about whether the label grabbed their attention and whether they read and looked closely at the label. Responses will be averaged to create a mean attention score. Responses are on a 1 to 5 scale. Thus, mean scores can range from 1-5, with higher scores indicating higher attention to the labels.
Time frame: Within 30 minutes following completion of ~10-minute shopping task
Mean Emotional Reactions Score to Trial Labels
Emotional reactions to the trial labels will be measured at post-test using 6 items adapted from Brewer et al. (2018). The items ask participants to indicate the extent to which trial labels elicited particular emotional reactions (e.g., fear, guilt, disgust). Responses will be averaged to create a mean emotional reactions score. Response options are on a 1 to 5 scale. Thus, mean scores can range from 1-5, with higher scores indicating stronger emotional reactions.
Time frame: Within 30 minutes following completion of ~10-minute shopping task
Mean Cognitive Elaboration Score
Cognitive elaboration will be measured at post-test using 2 items adapted from Moodie, MacKintosh \& Hammond (2010) and Fathelrahman et al. (2010). Items ask participants to rate the amount they thought about the information conveyed by the trial label and about the health consequences of consuming beverages with added sugar. Responses will be averaged to create a mean cognitive elaboration score. Response options range from 1 to 5. Thus, mean scores can range from 1-5, with higher scores indicating more elaboration.
Time frame: Within 30 minutes following completion of ~10-minute shopping task
Mean Social Interactions About the Labels
Social interactions of the labels will be measured at post-test with a single item adapted from Brewer et al. (2018). Participants will rate the extent to which they are likely to have social interactions about the trial label. Response options are on a 1 to 5 scale. Thus, mean scores can range from 1-5, with higher scores indicating a higher likelihood of having social interactions.
Time frame: Within 30 minutes following completion of ~10-minute shopping task
Mean Perceptions of Added Sugar Score Content in Specific Sugar-sweetened Beverages
Perceptions of added sugar content in specific categories of sugar-sweetened beverages (i.e., soda, fruit drinks, sport drinks/flavored waters, sweetened coffee/tea, and energy drinks) will be measured at post-test using items adapted from Roberto et al. (2016). Items ask participants to rate the amount of added sugar they think is in 1 serving of the beverage. Responses will be averaged to create a mean perceptions of added sugar score. Response options are on a 1 to 4 scale. Thus, mean scores can range from 1-4, with higher scores indicating perceiving higher added sugar content.
Time frame: Within 30 minutes following completion of ~10-minute shopping task
Mean Attitudes Score Toward Consuming Specific Sugar-sweetened Beverages
Attitudes toward consuming specific sugar-sweetened beverages (i.e., soda, fruit drinks, sport drinks/flavored waters, sweetened coffee/tea, and energy drinks) will be measured at post-test using 5 items adapted from Bollard et al. (2016). Items ask participants to rate how healthy/unhealthy it is for them to consume each beverage category. Responses will be averaged to create an average attitudes score. Response options are on a 1 to 7 scale. Thus, mean scores can range from 1-7, with higher scores indicating more positive attitudes toward consuming these sugar-sweetened beverages.
Time frame: Within 30 minutes following completion of ~10-minute shopping task
Mean Product Attitudes Score Toward Specific Sugar-sweetened Beverages
Product attitudes toward specific sugar-sweetened beverages (i.e., soda, fruit drinks, sport drinks/flavored waters, sweetened coffee/tea, and energy drinks) will be measured at post-test using 10 items adapted from Bollard et al. (2016). Items ask participants to rate the appeal and the coolness of each beverage category. Responses will be averaged to create a mean product attitude score. Responses are on a 1 to 7 scale. Thus, mean scores can range from 1-7, with higher scores indicating more positive product attitudes toward these sugar-sweetened beverages.
Time frame: Within 30 minutes following completion of ~10-minute shopping task
Mean Outcome Expectations Score Regarding Consumption of Beverages With Added Sugar
Outcome expectations will be measured at post-test with items adapted from Roberto et al. (2016). Participants will rate the extent to which they expect that consuming beverages with added sugar would increase their risk of negative health outcomes (e.g., tooth decay). Responses will be averaged to a create a mean outcome expectations score. Responses are on a 1 to 7 scale. Thus, mean scores can range from 1-7, with higher scores indicating more negative outcome expectations.
Time frame: Within 30 minutes following completion of ~10-minute shopping task
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.