FOOD INSECURITY AND FOOD DECISION-MAKING: USES OF BEHAVIORAL NUDGES IN A FOOD PANTRY SETTING

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: HATCH
• Reporting Frequency: Annual
• Accession No.: 1010557
• Project Start Date: Oct 11, 2016
• Project End Date: Sep 30, 2018
• Program Code: [(N/A)]- (N/A)

Recipient Organization
CORNELL UNIVERSITY
(N/A)
ITHACA,NY 14853

Performing Department
Applied Economics & Management

Non Technical Summary
According to the 2014 USDA's report on food insecurity in the United States the prevalence of food insecurity was 14% of U.S. households. Furthermore 19.5% of households with children were deemed food insecure. Many of these food insecure households are facing significant diet-related health challenges.Among these, obesity has been found to be closely linked to food insecurity and poverty.There is also evidence that those living in poverty are susceptible to scarcity-induced poorer quality in decision-making, from which it follows that there are few areas of a person's life upon which we would not expect food insecurity to have some undesirable effect.In view of these two empirical facts, behavioral economic approaches to address obesity and improve decision quality have appeal given their potential to effectively improve nutrition choices without patronizing or restricting consumers.Our overall goal is to find field-tested, scalable low- or no-cost behavioral interventions (nudges) that can lead food-insecure families to choose more nutritious foods to be implemented through Feeding America's network of hunger relief agencies. To accomplish this goal we will develop 10 low- or no-cost nudges and measure their efficacy by running a series of field experiments over the course of 12 weeks across 3 to 4 states in a variety of food pantry settings. For each of the field experiments we will define the targeted food and assign to it a nudge intervention based on one or more of the key following key behavioral principles: convenience, visibility, setting taste expectations, social norms, smart choice structure, and priming. We will distill key findings and successful nudges into a Practitioner's Guide to Nudging Nutrition that will summarize best practices and provide simple ways in which pantry-managers can implement nudges to increase selection of nutritious foods. This guide will be disseminated throughout the Feeding America network of 60,000 member agencies as well as made available publicly through their Healthy Food Bank Hub website: https://healthyfoodbankhub.feedingamerica.org/

Animal Health Component

25%

Research Effort Categories

Basic

75%

Applied

25%

Developmental

(N/A)

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
703	6010	3010	100%

Knowledge Area
703 - Nutrition Education and Behavior;

Subject Of Investigation
6010 - Individuals;

Field Of Science
3010 - Economics;

Keywords

food insecurity

food pantry setting

behavioral nudges

behavioral economic approaches to address obesity

obesity

low or no cost interventions

feeding america's foods to encourage

healthy choices in food pantries

food banks

Goals / Objectives
Our overall goal is to find field-tested, scale-able low- or no-cost behavioral interventions (nudges) that can lead food-insecure families to choose more nutritious foods to be implemented through Feeding America's network of hunger relief agencies.To accomplish this goal our specific objectives encompass:-To develop 10 low- or no-cost nudges and to measure their efficacy by running a series of field experiments over the course of 12 weeks across 3 to 4 states in a variety of food pantry settings.-For each of the field experiments we will define the targeted food and assign to it a nudge intervention based on one or more of the key following key behavioral principles: convenience, visibility, setting taste expectations, social norms, smart choice structure, and priming-Distill key findings and successful nudges into a Practitioner's Guide to Nudging Nutrition that will summarize best practices and provide simple ways in which pantry-managers can implement nudges to increase selection of nutritious foods. This guide will be disseminated throughout the Feeding America network of 60,000 member agencies as well as made available publicly through their Healthy Food Bank Hub website: https://healthyfoodbankhub.feedingamerica.org/-Report research findings through journal articles and issues policy recommendations to make them available to a wider audience.

Project Methods
For each of the 10 interventions, we will identify 6 food pantries in 3 different locations (control and treatment in each) in which to conduct field experiments. For each intervention, we will identify a specific Foods to Encourage (F2E) item, and work with Feeding America to ensure that there is sufficient supply of that food. Feeding America's recommended F2E framework encompasses products that more consistently meet the USDA 2010 Dietary Guidelines for Americans. Through their staff of Registered Dieticians Feeding America will assist in selecting those items that will be targeted with the behavioral nudges. Three of the pantries will be randomly assigned to a control group, and three will be assigned to the intervention group. The intervention will last 6 weeks. Our research team will collect 6 weeks of baseline data and 6 weeks of data post intervention. Each day of data collection will consist of recording the selection of the targeted food by each patron. We have found that accurate data can be obtained from a single observer.This keeps expenses required for data collection to a minimum.We will measure the effects of the behavioral interventions by observing the selection of food items from food pantries in both treatment and control groups. A comparison of changes in selection from the two groups before and after the treatment period will be used to determine intervention effectiveness using a difference-in-differences estimation strategy which allows us to control for a food-choice-related factors (including things like seasonality of a given food item and all other factors common to both treated and control groups across time). This empirical approach helps to ensure the accuracy of the measured nudge treatment effects.These data, combined with detailed observations about any challenges encountered with implementation, will be used to compile a set of best practices in using nudges to encourage selection of nutritious foods in the hunger relief agency setting. We will work with Feeding America to refine these best practices which will be compiled into the Practitioner's Guide to Nudging Nutrition for use by food pantries. Feeding America will make these findings available to their network of over 60,000 hunger relief agencies and to the public through their online resources.

Progress 10/11/16 to 09/30/18

Outputs
Target Audience:Our efforts to reach the audience of food pantry managers has encompassed two distinct efforts. First, we have worked with local food pantries in the Ithaca and Horseheads area. In these efforts we have disseminated some of our early findings on convenience, value and choice in the food pantry environment. These include results that were found through experiments in the local food pantries themselves. Pantry managers have expressed interest in continuing this relationship to help improve their operations. We have two distinct studies planned for after the close of this Hatch grant that will help us to maintain this relationship. In addition, our efforts have led to similar relationships withthe Meals on Wheels program where we have some studies planned for the coming year. Secondly, we continued to work with Feeding America for much of this reporting year as they began to scale up their behavioral economic efforts more robustly and more broadly. This included providing some feedback on their materials to be distributed to pantries. Their training materials are distributed to the majority of food pantries in the nation.We had also planned continued studies collaborating with them in the context of school backpack programs. However, the status of this project is unclear and may not come to fruition. We had been at the point of selecting target locations, but delays have set back this project and it may have become a low priority for Feeding America. Changes/Problems:The employment status of one of the collaborators changed significantly over the course of the project. This led to several complications in the publication of results and some complications in research relationships with organizations and institutions. This did not lead to a material change in the proposed work, but had a significant impact on dissemination of results to the academic community. What opportunities for training and professional development has the project provided?While we did not intend to provide direct professional development opportunities, our collaboration with Feeding America has led to the training of many pantry managers nationwide in the use of behavioral techniques to encourage healthy food consumption. How have the results been disseminated to communities of interest?The research team has made presentations to several food banks and pantries both in our local area and in areas where studies were conducted (Oklahoma, Colorado, New Jersey, Texas). In addition, we have helped prepare materials for dissemination by Feeding America through their network of 60,000 pantries nationwide. These materials include web guides, reports on results of the studies and printed guides on how to implement behavioral interventions. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? In our work with food pantries we were able to develop and test many specific low cost nudges in actual food pantry contexts. In conducting this test we quickly realized that some nudges will be highly context specific. Our workyielded some nudges that appear to have significant impacts on selection of healthier foods. These included: 1) ensuring that produce is presented at waist level rather than at ground level, 2) signage that highlights the hedonic properties of foods rather than nutrition information, 3) the use of overhead signage for fresh fruit that can be seen over food aisles, 4) reducing the apparent abundance of less healthy foods by only displaying a few examples and storing the remainder out of sight, 5) displaying whole grain items in more than one location on a line, 6) placing healthier foods in the front of the line, 7) retaining commercial quality packaging when possible. These results were helpful in leading to reports issued by Feeding America and distributed nationwide providing something of a practitioners guide to using nudges. Two of these nudges were published in an article appearing in the Journal of Public Health. Two other nudges were included in an article that was accepted for publication in a leading journal, and then withdrawn over disputes among the co-authors of the study. It is still hoped that these disputes can be resolved so that these findings and the remaining findings can make their way into the published literature.

Publications

Progress 10/11/16 to 09/30/17

Outputs
Target Audience:The audience for this research are managers and other decision-makers in the emergency food assistance realm. Working with Feeding America, our research has been presented to those managing food banks that supply approximatelyhalf of the food pantries in the U.S. We have worked with them toadapt our materials for dissemination to this audience. In addition to this, we have presented our results at several nutrition and dietetics conferences to other professionals that work in this space. Built upon our results, Feeding America has just recently committed to rolling out a behavioral economics based program to their entire network. Changes/Problems:As mentioned, we found that some of the nudge interventions we had anticipated would be impractical due to management considerations of most pantries. For this reason, while we continue to work with pantries on the remaining nudges, we are also considering other behavioral interventions that include reframing food assistance under the banner of medical care. In addition to the normal research challenges, the Co-PI on the project has faced significant public controversy. This has made it difficult to proceed with publication of a portion of the already completed research. In particular, an interpersonal issue led to the withdrawal of an already accepted paper. I am working to ensure such issues will not stand in the way of future publication or research. What opportunities for training and professional development has the project provided?The grant has been helpful in training a PhD student on the conduct of field experiments as well as working with professional assistance organizations. How have the results been disseminated to communities of interest?We have relied on Feeding America to help disseminate our materials to those who make managerial decisions in food assistance organizations. This includes several different reports and training materials that they have helped to develop. Moreover, they are now working on a much larger program based upon our work that will inform their entire network of food banks. What do you plan to do during the next reporting period to accomplish the goals?We are working to design and carry out experiments both with online food bank ordering platforms and school backpack programs. It is anticipated that the backpack program experiments will be conducted in Minnesota, though we are still in negotiation. In addition, we are continuing work with the food bank in Colorado to explore the impact of travel time on patrons as well as the potential to pair food pantry services with basic medical care to reduce both hunger and other health issues. The travel time project will use pantry data on patrons as well as generally available secondary data. This work is already proceeding and we anticipate that a draft of the research will be completed within the reporting period. We are currently designing the study of pairing medical care with food assistance. We anticipate that this study will be under way within the reporting period.

Impacts
What was accomplished under these goals? We have conducted field experiments in food pantries in New York, New Jersey, Colorado, Oklahoma and Texas. This includes the testing of 7 distinct nudges, which have been classified in to 3 different nudge types: visibility, convenience and messaging. This work is in preparation for publication. In working with food pantries we found some of our original ideas to be impractical and have dropped them from consideration. We are continuing experiments with food bank level nudging via online ordering platforms and the use of backpack programs for influencing decisions.

Publications

• Type: Journal Articles Status: Published Year Published: 2016 Citation: Wilson, Norbert LW, et al. "Food pantry selection solutions: a randomized controlled trial in client-choice food pantries to nudge clients to targeted foods." Journal of Public Health 39.2 (2016): 366-372.