CTSI Pilot: Improving Adherence to Diabetic Diet

CTSI Pilot: Improving Adherence to a Diabetic Diet With a Grocery Shopping Intervention

425 million adults live with diabetes worldwide, and the prevalence of Type 2 diabetes is rising. Dietary approaches are recommended for weight control and diabetes management, but modern environments, characterized by plentiful, unhealthy foods, pose challenges to selecting a healthy diet. Behavioral economics offers a framework for modifying the food environment to encourage individuals with diabetes to select low-calorie and low-sugar foods. The goal of this study is to test novel approaches informed by behavioral economics to promote healthier grocery shopping among diabetic patients. Adults who have Type 2 diabetes or who are at risk for developing Type 2 diabetes will be recruited. Participants will be randomly assigned to 1 of 2 interventions or a control group in which they will shop in-person as usual. The Online intervention will utilize online grocery shopping to promote healthier purchasing. The Defaults intervention will augment this intervention, showing participants a default shopping cart pre-filled with items that correspond to the DASH diet and diabetic diet goals, which they may modify as they like. Receipt data will be collected to quantify the alignment of purchases with diabetic diet goals before, during, and after interventions. Purchases lower in calories, carbohydrates, and sugar and higher in nutritional quality (DASH diet score) are expected in the Defaults group; the Online group is expected to have intermediary results between Defaults and Controls. The investigators will also explore effects of the interventions on spending and dietary intake. This study is intended to demonstrate the efficacy of strategies that leverage behavioral economics principles to make the purchasing of healthier foods easier. The strategies have translational significance as they could be incorporated into clinical treatment, with the potential to improve dietary intake, glucose regulation, weight, and medication needs among diabetic patients.

Study Overview

• Status: Completed
• Conditions: Diabetes Mellitus, Type 2; Diet, Healthy; Pre-diabetes
• Intervention / Treatment: Behavioral: Online - use of online grocery shopping platform; Behavioral: Default - use of online grocery shopping platform with default shopping carts

Detailed Description

The goal of this study is to test the effects of interventions informed by behavioral economics on the healthfulness of grocery purchases among diabetic adults. This study has a 3-group experimental design; adult patients with diabetes or at risk for diabetes who typically grocery shop in-person will be randomly assigned to one of two interventions or a control group. In all groups, recipe cards that follow the evidence-based DASH diet and correspond to dietary recommendations for diabetic patients will be provided. In addition, the first intervention, Online, will utilize online grocery shopping at a local grocery store to promote healthier purchasing, with a goal of removing visceral factors that can lead to impulse purchases in stores. The second intervention, Defaults, will augment the Online intervention, showing participants a default shopping cart when they log into their accounts. They will be told that their cart has been filled with items that conform to a diabetic diet and can be used to make the recipes from the provided recipe cards, and that they may modify their carts as they like. In the control group, participants will shop in-person as usual.

At baseline, all participants will complete a survey online or via phone and then will grocery shop in-person as usual. Participants will then be randomized and receive further instructions. For the next 3 weeks, they will receive recipe cards and shop in accordance with their assigned study group. The following week (post), they will grocery shop using the mode of their choice (in-person or online) and complete a survey again. Household receipt data will be collected throughout the study. The baseline period will also serve as a run-in period, and participants whose diabetes/prediabetes status is not confirmed by a physician or who do not comply with study guidelines for baseline procedures will not be randomized for further participation. Primary analyses will examine intervention effects on grocery purchases, with additional analyses examining spending and dietary intake. All participant interactions will occur online or via phone, with cards for participant payment sent via postal mail.

The specific aims of this study are: (1) to test effects of Online and Defaults interventions on diabetic adults' grocery purchases and (2) to explore effects of the Online and Defaults interventions on participants' spending during grocery shopping. It is hypothesized that (1a) purchases made by adults in the Defaults group will be lower in calories, carbohydrates, and sugar and higher in nutritional quality (DASH diet score) at post-test versus other study groups, (1b) adults in the Defaults group will show the greatest increases in nutritional quality versus other study groups, (1c) the Online group will have intermediary results between Defaults and Controls, and (2) there will be no difference in total dollars depicted on receipts across study groups during the intervention period (i.e. the costs of online shopping will be offset by effects of the Online and Default interventions).

• Study Type: Interventional
• Enrollment (Actual): 65
• Phase: Not Applicable

Contacts and Locations

Study Locations

• United States
  • New York
    • Buffalo, New York, United States, 14214
      • State University of New York at Buffalo

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years and older (Adult, Older Adult)
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

• Must be at least 18 years old
• Must have been diagnosed diabetes or demonstrate diabetes risk as defined by the American Diabetes Association
• Must shop in-person at select local grocery stores at least weekly, and at least 75% of the grocery shopping for the household must be done at these stores
• Must do at least 75% of the grocery shopping for the household
• Must speak English

Exclusion Criteria:

• Is under 18 years of age
• Does not meet diabetes/risk for diabetes criteria
• Does not shop in-person at select local grocery stores at least weekly
• Does not live in a household in which 75% of groceries come from select local grocery stores
• Does not do at least 75% of the grocery shopping for the household
• Does not speak English
• Receives SNAP or WIC benefits (an exclusion criterion for this study due to incompatibility of EBT cards with the selected online grocery shopping platform at this time)
• Has dietary restrictions or preferences that would not allow them to reasonably partake in the study (i.e. they would not be willing or able to buy/eat many of the staple foods included in default carts/recipe cards)

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Other
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Single

Number of Arms

3

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
No Intervention: Group 1 - Control; No intervention - participants will shop in-person as they usually would. Participants will receive recipe cards that follow the evidence-based DASH diet and correspond to dietary recommendations for diabetic patients.
Experimental: Group 2 - Online; Participants will utilize online grocery shopping (shopping at a local grocery store via an online platform). Participants will also receive recipe cards that follow the evidence-based DASH diet and correspond to dietary recommendations for diabetic patients.	Behavioral: Online - use of online grocery shopping platform; The Online intervention will utilize online grocery shopping (shopping at a local grocery store via online platform).
Experimental: Group 3 - Default; The default intervention will augment the Online intervention, showing participants a default cart when they log into their online grocery shopping accounts. They will be told that their cart has been filled with items that conform to a diabetic diet and can be used to make recipes from the provided recipe cards, and that they can modify it as they like. Participants will also receive recipe cards that follow the evidence-based DASH diet and correspond to dietary recommendations for diabetic patients.	Behavioral: Default - use of online grocery shopping platform with default shopping carts; The Defaults intervention, will augment the Online intervention, showing participants a default cart when they log into their online grocery shopping accounts. They will be told that their cart has been filled with items that conform to a diabetic diet and can be used to make the recipes from provided recipe cards, and that they may modify their cart as they like.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Calories Purchased	Receipt data will be entered into Nutritionist Pro software and will be used to calculate total calories purchased from target grocery stores.	At post-test (week 5)
Calories Purchased	Receipt data are entered into Nutritionist Pro software and used to calculate total calories purchased from target grocery stores each week. For this analysis, the weekly values were aggregated, such that the least squares means reported herein represent average calories purchased across all 5 weeks, in the context of the repeated measures ANOVA model.	Weeks 1-5 (testing overall main effect of study group)
Change in Calories Purchased	Receipt data will be entered into Nutritionist Pro and will be used to calculate changes in total calories purchased from target grocery stores.	Baseline (week 1), intervention (weeks 2-4), and post-test (week 5)
Carbohydrates Purchased	Receipt data will be entered into Nutritionist Pro and will be used to calculate carbohydrates purchased from target grocery stores.	At post-test (week 5)
Carbohydrates Purchased	Receipt data are entered into Nutritionist Pro and used to calculate carbohydrates purchased from target grocery stores each week. For this analysis, the weekly values were aggregated, such that the least squares means reported herein represent average carbohydrates (grams) purchased across all 5 weeks, in the context of the repeated measures ANOVA model.	Weeks 1-5 (testing overall main effect of study group)
Change in Carbohydrates Purchased	Receipt data will be entered into Nutritionist Pro and will be used to calculate changes in carbohydrates purchased from target grocery stores.	Baseline (week 1), intervention (weeks 2-4), and post-test (week 5)
Sugars Purchased	Receipt data will be entered into Nutritionist Pro and will be used to calculate sugars purchased from target grocery stores.	At post-test (week 5)
Sugars Purchased	Receipt data are entered into Nutritionist Pro software and used to calculate total sugar purchased from target grocery stores each week. For this analysis, the weekly values were aggregated, such that the least squares means reported herein represent average sugar (grams) purchased across all 5 weeks, in the context of the repeated measures ANOVA model.	Weeks 1-5 (testing overall main effect of study group)
Change in Sugars Purchased	Receipt data will be entered into Nutritionist Pro and will be used to calculate changes in sugars purchased from target grocery stores.	Baseline (week 1), intervention (weeks 2-4), and post-test (week 5)
Nutritional Quality of Purchases	Nutritional quality of grocery purchases was created using indicators adapted from Fung 2008: fruit, vegetables, nuts/legumes, whole grains, low-fat dairy, sodium, red/processed meat, carbohydrates. Quintiles were calculated for each, using the number of items from a food group or sodium [mg] and carbohydrates [g] purchased that week. Resulting scores from 1-5 indicate whether participants were lower or higher on each indicator (red meat, sodium, carbohydrates reverse scored). The 8 scores were summed for a total nutritional quality score ranging from 8 to 40 (40=highest nutritional quality).	At post-test (week 5)
Nutritional Quality of Purchases	Nutritional quality of grocery purchases was created using indicators adapted from Fung 2008: fruit, vegetables, nuts/legumes, whole grains, low-fat dairy, sodium, red/processed meat, carbohydrates. Quintiles were calculated for each, using the number of items from a food group or sodium [mg] and carbohydrates [g] purchased that week. Resulting scores from 1-5 indicate whether participants were lower or higher on each indicator (red meat, sodium, carbohydrates reverse scored). The 8 scores were summed for a total nutritional quality score ranging from 8 to 40 (40=highest nutritional quality).	Weeks 1-5 (testing overall main effect of study group)
Change in Nutritional Quality of Purchases	Nutritional quality of grocery purchases was created using indicators adapted from Fung 2008: fruit, vegetables, nuts/legumes, whole grains, low-fat dairy, sodium, red/processed meat, carbohydrates. Quintiles were calculated for each, using the number of items from a food group or sodium [mg] and carbohydrates [g] purchased that week. Resulting scores from 1-5 indicate whether participants were lower or higher on each indicator (red meat, sodium, carbohydrates reverse scored). The 8 scores were summed for a total nutritional quality score ranging from 8 to 40 (40=highest nutritional quality)	Baseline (week 1), intervention (weeks 2-4), and post-test (week 5)

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Dollars Spent	Total dollars spent at the target grocery stores from submitted receipts	At post-test (week 5)
Dollars Spent	Total dollars spent at the target grocery stores from submitted receipts	Weeks 1-5 (testing overall main effect of study group)
Change in Dollars Spent	Changes in total dollars spent at the target grocery stores from submitted receipts	Baseline (week 1), intervention (weeks 2-4), and post-test (week 5)
Nutritional Quality of Foods Consumed	Nutritional quality of grocery purchases was created from food frequency questionnaire data using indicators adapted from Fung 2008: fruit, vegetables, nuts/legumes, whole grains, low-fat dairy, sodium, red/processed meat, carbohydrates. Quintiles were calculated for each, using reports of the frequency that the food group or nutrient of interest was consumed over the past month. Resulting scores from 1-5 indicate whether participants were lower or higher on each indicator (red meat, sodium, carbohydrates reverse scored). The 8 scores were summed for a total nutritional quality score ranging from 8 to 40 (40=highest nutritional quality).	At post-test (week 5)

Collaborators and Investigators

• Sponsor: State University of New York at Buffalo
• Collaborators: National Center for Advancing Translational Sciences (NCATS)
• Investigators: Principal Investigator: Stephanie Anzman Frasca, PhD, Associate Professor

Study record dates

Study Major Dates

• Study Start (Actual): October 14, 2019
• Primary Completion (Actual): December 3, 2020
• Study Completion (Actual): December 3, 2020

Study Registration Dates

• First Submitted: July 31, 2019
• First Submitted That Met QC Criteria: August 6, 2019
• First Posted (Actual): August 9, 2019

Study Record Updates

• Last Update Posted (Estimated): December 13, 2023
• Last Update Submitted That Met QC Criteria: November 20, 2023
• Last Verified: November 1, 2023

More Information

Terms related to this study

• Keywords: diabetes; diet; grocery shopping; online shopping; behavioral economics
• Additional Relevant MeSH Terms: Glucose Metabolism Disorders; Metabolic Diseases; Endocrine System Diseases; Diabetes Mellitus; Diabetes Mellitus, Type 2; Prediabetic State
• Other Study ID Numbers: STUDY00003054; UL1TR001412 (U.S. NIH Grant/Contract)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No