2016 Special Solicitation WIC Research Grants

2016 Special Solicitation WIC Grants: Improving the WIC Shopping Experience Using Behavioral Economics-Based Approaches

We recently funded 3 proposals that draw on behavioral economics theory to develop and test strategies for improving food choice behaviors in the WIC shopping context. These projects run from January 1, 2017 to December 31, 2017.

Improving Food Choices for Low-income Families with Young Children: A Supermarket Intervention

PI: Lydia Ashton, University of Wisconsin-Madison

Background: Using   information   technology   to   improve   the   Special   Supplemental   Nutrition Program   for   Women,   Infants   and   Children   (WIC)   service   delivery   and   program management  has  been  a  priority  for  several  years.  While  some  progress  has  been made  in  this  area,  there  is  room  for  much  more.  Barriers  in  procuring  food  are  often identified  as  one  of  the  top  barriers  to  the  use  of  WIC  services.  These  barriers  may include problems due to product availability and/or identification. Evidence suggests that choice  overload  may  play  a  significant  role  in  deterring  consumers  from  purchasing decisions. Another barrier in procuring food that WIC participants face is the lack of time (e.g. due to employment).

Specific Aims: We  will  test  whether  focusing  consumers’  attention  through  an  online  WIC-allowed feature, which will filter products according to their WIC eligibility for customers placing grocery   orders   online,   would   facilitate   food   procurement   and   increase   benefit redemption (Aim 1). We will also test whether the online ordering system can reduce the time  that  the  WIC  participants  spend  in  the  grocery  store  and  improve  WIC  participant retention (Aim 2). Lastly, we will also test whether a pre-ordering system, i.e. scheduling weekly  or  bi-weekly  pickups,  can  improve  redemption  rates  and  help  participants  to smooth their consumption throughout the benefit period (Aim 3).

Research Design and Methods: In collaboration with a national retail grocery  store  partner, we will design a web/mobile app that will allow us to conduct a randomized controlled experiment. We will recruit 450 WIC and non-WIC customers. Participants will be randomly assigned to one of the conditions. We will track participants’ purchasing decisions and behaviors for a  period  of  at  least  1 month  before and 6 months after the experiment, which will last between 4 and 12 weeks. We will track redemption rates for WIC participants for in-store and online customers. This key variable will provide us with a real-time indication of participation, which is our first outcome of interest. In addition, we will collect data, via online or mobile surveys; on whether participants attended the next 2 WIC visits that fall within the 6 month post-experiment time frame (nutrition ed appointment, mid-cert, or sub-cert, depending on what the mother is due for). We will also ask whether there was a  gap,  measured  in  days, between their benefit periods. This will allow us to obtain a measure of retention, which is our second outcome of interest. The experimental design will consist of: one control or on-demand condition, where customers will be able to place an order at any time; treatment 1 or on-demand with WIC-allowed  identification feature  condition,  where  customers  can  place  an order at any time and the WIC-allowed feature will be available to them; treatment 2 or meal-plan condition, where customers are prompted with the option of scheduling weekly orders during the first or second week of the month; and treatment 3 or meal-plan with WIC-allowed feature condition, where customers are prompted with the option of scheduling weekly orders during the first or second week of the month and the WIC-allowed feature will be available to them.


Identifying optimal behavioral economic strategies to increase WIC redemptions by low income women in Baltimore corner stores

PI: Joel Gittelsohn, Johns Hopkins Global Obesity Prevention Center

Background: Low-income US populations face a dual burden of food insecurity and obesity. In Baltimore City, 79% of food insecure adults are also obese. Furthermore, 25% of residents live in a food desert where corner stores are a primary food source, making nutrient poor, convenient, and inexpensive food the status quo. In Baltimore’s food deserts, small stores that offer WIC would appear to offer a partial solution to this problem. However, there was a 5% decline in WIC participation in Baltimore in just the past year. In our own work, we have identified numerous barriers to selling WIC foods, including: perceived lack of demand for WIC foods, challenges in stocking perishable foods, language barriers, embarrassment at checkout, and difficulty finding WIC products – which deter participants from taking full advantage of the WIC program. Behavioral economic strategies have been used in public health promotions to shape consumer behavior and make healthier choices more attractive and salient. However, few studies have evaluated use of these strategies in small urban food stores.

Specific Aim: The overall objective of the proposed study is to test, alone and in combination, multiple behavioral economic strategies in small corner stores, and determine their impact on stocking and sales of WIC foods, as well as on WIC customer purchase of these foods.  We aim to determine the impact of: 1) four different behavioral economic strategies separately on stocking and sales of WIC foods; 2) combined behavioral economics strategies on stocking and sales of WIC foods; 3) the different behavioral economics strategies, alone and in combination, on consumer selection and purchasing of WIC foods.

Research design: The proposed study uses a randomized controlled trial design, where 10 corner stores will be randomized after enrollment in the study to one of the five treatment groups (intervention=8, comparison=2). Aspects of the social ecological model, social cognitive theory and behavioral economics nudging strategies guide this study. We will test four different intervention strategies: A) Store owner training. Storeowners will watch short videos focused on promotion of WIC foods through nudging and framing, B) Enhanced shelf labeling and posters to identify WIC-approved foods and their health benefits, C) Product placement (i.e. at eye level or register), and D) Grouping WIC products together within the store (i.e. endcap displays). Treatments will be combined in successive months. The final treatment month will combine all four intervention treatments to assess collective effects. The sequence of testing interventions and comparison group will allow us to tease out the impact of specific interventions alone and in combination with others, as well as the effect of the order of treatments. We will assess changes in stocking, sales, and purchasing of WIC foods to determine the impact of the intervention.  The study provides a unique opportunity to assess which combination of low cost behavioral economic strategies leads to greatest sales and purchasing of WIC foods in a low income urban African American corner stores.


Using a Smartphone App to Improve WIC Redemptions

PI : Qi (Harry) Zhang, Old Dominion University Foundation

Background: The USDA requires all WIC agencies to implement the Electronic Benefit Transfer (EBT) system by October 1, 2020 (USDA, 2016). Virginia WIC implemented the EBT system statewide in May 2014. In the ongoing BECR Center-funded project, we conducted focus groups with Virginia WIC participants about their experience with the EBT system. The results indicate that although the system improves the efficiency of WIC transactions, significant behavioral barriers still exist when participants redeem their WIC benefits, particularly in certain food categories such as breakfast cereal (“cereal” hereafter) and fruits and vegetables (F&V). Not all cereal brands are eligible for WIC,and the cereal benefit is capped by total weight in ounces. Due to the various brand names and package sizes, the mental accounting capabilities needed to adjust for these variations may prevent participants from redeeming their full cereal benefits. F&V benefits, on the other hand,are capped by cash value. Given the diverse variety of F&V and various unit prices, participants have to calculate the cost of F&V carefully so as not to exceeding the cap value. With these barriers, participants are less likely to deviate from their default behavioral patterns and to redeem their full WIC benefits. Therefore, it is important to identify effective interventions to facilitate their redemption experience.

Specific Aims: The overall goal of this pilot study is to test whether a smartphone app with an “adaptive calculator” can reduce behavioral barriers and improve the redemption rates of cereal and F&V among Virginia WIC participants. The specific aims are: Aim 1: To add an “adaptive calculator” module in a WIC app to make cereal and F&V redemption easier; Aim 2: To conduct a field experiment implementing the customized app in Virginia and evaluate the app’s influence on participants’ cereal and F&V redemptions.

Research Design and Methods: This pilot project will be carried out in three stages. Stage 1: App customization: We will customize the app, WICShopper®, based on the requirements of the Virginia WIC program and add an “adaptive calculator” module to the app. The adaptive calculator is designed to help participants calculate their remaining balances and develop potentially viable cereal or F&V redemption plans. Stage 2: App implementation: We will work with the Virginia WIC agency and JPMA, Inc., the development firm of WICShopper®,to integrate the app’s database with the WIC administrative database so the app can verify the eligibility of food products based on the daily updated approved product list (APL). Stage 3: Field experiment: We will use a clustered pretest-posttest control group experimental design to test the effect of the app on participants’ redemption rates for cereal and F&V. We target to recruit approximately 1,000 participants (50 each in 20 cities/counties in Virginia) and to randomize 10 cities/counties into the intervention group and the other 10 cities/counties into the control group. The intervention group will have the app login information,while the control group will not. The app will be used for two months. We will survey participants’redemption experiences and WIC program satisfaction both before and after the intervention. We will compare the two groups’ redemption rates in cereal and F&V and also the intervention group’s redemption rates pre-and post-use of the app. We will follow up with the intervention groups about their experiences with the app and solicit their responses on how to improve it. In summary, through this study, we will better understand whether an app can address behavioral barriers and help improve participants’ redemption of WIC benefits, particularly for cereal and F&V. If successful, the Virginia WIC agency will be likely to adopt the WIC app in the state.