x

We are happy to announce that all trial registrations will now be issued DOIs (digital object identifiers). For more information, see here.
Smart Shopping Algorithm (SSA) to Improve Nutrition
Last registered on August 09, 2019

Pre-Trial

Trial Information
General Information
Title
Smart Shopping Algorithm (SSA) to Improve Nutrition
RCT ID
AEARCTR-0004520
Initial registration date
August 08, 2019
Last updated
August 09, 2019 9:43 AM EDT
Location(s)

This section is unavailable to the public. Use the button below to request access to this information.

Request Information
Primary Investigator
Affiliation
Duke-NUS Medical School
Other Primary Investigator(s)
Additional Trial Information
Status
In development
Start date
2019-08-23
End date
2019-12-31
Secondary IDs
Abstract
While there have been efforts to improve the availability and transparency of nutritional information in food products through the provision of Nutritional Fact Panels (NFPs), they might not be immediately useful to the casual shopper.

In most standard-format NFPs, the information provided can be overwhelming and underused by the consumer for various reasons including a lack of time, a lack of nutritional literacy, the lack of a comprehensive denomination, or the unavailability of other baseline information (such as Daily Recommended Intake for different nutrients). This is further confounded when consumers wish to compare between products to make healthier decisions.

The rise in the popularity of online grocery stores is an opportunity to test the effectiveness of comprehensive nutrition tools that can provide consumers with nutritional information in clearer formats that can be better acted upon, overcoming the traditional barriers to healthier food decision-making.

This study seeks to assess the effectiveness of a Smart Shopping Algorithm (SSA) as an add-on to standard web-based grocery stores. The SSA is a suite of features implemented into a web based grocery store with the aim of providing a more convenient and nutrition-focused way of shopping for groceries.

We propose to conduct a two arm cross-over randomized controlled trial (RCT) on an experimental fully functioning web-based grocery store “NUSMart” to assess the effectiveness the SSA. Participants will be randomized into one of two intervention sequences, SSA then control, or control then SSA. In each arm, participants will perform one grocery shopping trip in that version of NUSMart; participants will be required to complete two shops to complete the study. In each shop, the nutritional data of their finalized purchases will be collected.

We hypothesize that diet quality, as measured by weight average Nutri-Score, will be significantly higher in the SSA arm than the control arm. Post-study surveys will collect participants’ experience and feedback for potential future extensions of the SSA.
External Link(s)
Registration Citation
Citation
Finkelstein, Eric . 2019. "Smart Shopping Algorithm (SSA) to Improve Nutrition." AEA RCT Registry. August 09. https://doi.org/10.1257/rct.4520-1.0.
Former Citation
Finkelstein, Eric . 2019. "Smart Shopping Algorithm (SSA) to Improve Nutrition." AEA RCT Registry. August 09. https://www.socialscienceregistry.org/trials/4520/history/51482.
Sponsors & Partners

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information
Experimental Details
Interventions
Intervention(s)
A two-arm cross-over randomized controlled trial (RCT) will be conducted on a web-based grocery store “NUSMart” to assess the effectiveness of an implemented Smart Shopping Algorithm (SSA) on consumers’ diet quality.

The results of this analysis will be complemented with participants’ baseline questionnaire and post-study survey to examine the causal effect of the SSA on food purchase decisions.

Participants will be randomized into one of two intervention sequences: Control condition followed by Intervention condition(SSA), or Intervention condition followed by Control condition.

Participants will be instructed to perform two grocery shopping trips on NUSMart, purchasing a week’s worth of groceries each time.

Data of their finalized purchases will be collected for each shopping trip via their sales orders for subsequent analyses.
Intervention Start Date
2019-08-23
Intervention End Date
2019-12-31
Primary Outcomes
Primary Outcomes (end points)
1. Diet quality per shopping trip, which will be calculated as the weighted average of all purchased products’ Nutri-Score for the shopping trip. Nutri-Score is an individual dietary index based on the British Food Standard Agency Nutrient Profiling System.
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
1. Diet quality per shopping trip, as calculated as the portion-per-day AHEI-2010 score.
2. Diet quality per shopping trip, as measured by the Grocery Purchase Quality Index-2016 (GPQI-2016). The GPQI is based on USDA Food Plan expenditure shares and includes 10 food-based components of the HEI-2010, plus processed meats.
3. Average calories, sodium, sugar, fat, and saturated fat per Shopping Trip, which will be calculated as the sum of all purchased products’ total nutritional value for the shopping trip.
4. Calorie per dollar (kcal per $) will be calculated since labels may simultaneously influence diet quality, cost and purchase volume.
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
For this two-arm cross-over RCT, the experimental web-based grocery store “NUSMart” will be used. NUSMart was designed to mirror an actual web-based grocery store in look and feel. All products include pictures of the item, retail prices and product descriptions.

Participants will only be recruited if they are Singapore residents aged 21 years or older.

Participants will be randomized into one of two intervention sequences: Control condition followed by Intervention condition(SSA), or Intervention condition followed by Control condition.

Participants will be instructed to perform two grocery shopping trips on NUSMart, purchasing a week’s worth of groceries each time.

Data of their finalized purchases will be collected for each shopping trip via their sales orders for subsequent analyses.
Experimental Design Details
Not available
Randomization Method
Randomization done in office by a computer.
Randomization Unit
Individual.
Was the treatment clustered?
No
Experiment Characteristics
Sample size: planned number of clusters
136 individuals.
Sample size: planned number of observations
136 individuals.
Sample size (or number of clusters) by treatment arms
68 individuals in Sequence 1.
68 individuals in Sequence 2.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
We have calculated the sample size of a two-treatment cross-over design study. At 0.05 significance level, 0.8 statistical power, and an effect size of 0.29 for mean Nutri-score, it is calculated that we will need a sample size of 121. Including a 10% attrition (based on prior studies of similar nature), the sample size for this study is 136.
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
National University of Singapore-Institutional Review Board
IRB Approval Date
2019-07-25
IRB Approval Number
S-19-154