Nudging for Behavior Change in School Cafeterias

Measuring the Long-Term Impact of Behavioral Interventions in Middle School Cafeterias

The main purpose of this research is to look at a school lunchroom intervention that is known to improve fruit, vegetable, and milk consumption and see how it changes in the long run and if it affects the child's behavior permanently. The intervention will include the giving the vegetables descriptive names, moving the fruit to right next to the register and in attractive bowls, and increasing the amount of white milk served by 10%. The main forms of analyzing these results are through food preparation records, lunch sales records, and tray waste records. The first focus of this study is to see if there is a specific point in an intervention when improvement stops and therefore the intervention needs to be updated. This procedure involves looking at five similar middle schools with this same intervention over a 15 week period. The time of intervention implementation will vary by three week intervals, so the first will start the first week of school, the next school will start after three weeks of school, the next will start after six weeks of school, and the next will start after nine weeks of school, and the last will be a control school where there will be no intervention implementation. These intervals will help eliminate bias dealing with the beginning of the year excitement and seasonal effects. Food preparation records and lunch sales records will be collected from the school for the 15 week period. Tray waste will be recorded by having 200 randomly selected trays measured and collected twice a week over the 15 week period. This focus will help schools manage when they need to change their intervention so that improvements will not stop. The second focus of this study is to see if the children's improvement is kept when the intervention has stopped. This procedure will involve looking at a similar school to the other 5 schools. But unlike the other schools, this one will have the first 5 weeks without the intervention, then 5 weeks with the intervention, and then another 5 weeks without the intervention. Food preparation records and lunch sales records will be collected from the school for the 15 week period. Tray waste will be recorded by having 200 randomly selected trays measured and collected twice a week over the 15 week period. This second focus will help identify how effective this intervention is in permanently changing dietary habits.

Study Overview

• Status: Completed
• Conditions: Food Selection; Behavior, Child
• Intervention / Treatment: Behavioral: Basic School Lunchroom Nudges

• Study Type: Interventional
• Enrollment (Actual): 3000
• Phase: Not Applicable

Contacts and Locations

Study Locations

• United States
  • Ohio
    • Columbus, Ohio, United States, 43203
      • Champion Middle School
    • Columbus, Ohio, United States, 43204
      • Hilltonia Middle School
    • Columbus, Ohio, United States, 43207
      • Buckeye Middle School
    • Columbus, Ohio, United States, 43219
      • Mifflin Middle School
    • Columbus, Ohio, United States, 43224
      • Medina Middle School
    • Columbus, Ohio, United States, 43227
      • Johnson Park Middle School

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 10 years to 15 years (Child)
• Accepts Healthy Volunteers: Yes
• Genders Eligible for Study: All

Description

Inclusion Criteria:

• This research focused on food choices of middle school students, grades 6-8, which generally range in ages 11-14. In addition, since we are interested in learning about the types of cafeteria foods children select and what they throw away, we focus on students who receive a school lunch meal.

Exclusion Criteria:

• We exclude students who do not receive a school lunch meal.

Study Plan

How is the study designed?

Design Details

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

Number of Arms

6

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
No Intervention: Control; In this school, we collected data throughout the entire study without implementing any intervention.
Experimental: 5-week Intervention with Post-intervention Data Collection; In this school, we collected baseline data for 5 weeks, implemented the intervention for five weeks, then removed the intervention and collected post-intervention data for five weeks.	Behavioral: Basic School Lunchroom Nudges; The intervention consists of three components: 1) serving fresh fruit in attractive bowls in at least two locations on the lunch line; 2) providing descriptive names for the vegetables; and 3) increasing amount of plain milk availability by 10% and placing it in front of the chocolate milk. For the vegetable naming intervention, researchers first developed, printed, and laminated descriptive vegetable name cards (2"x4"). Some examples included "Savory Collard Greens," and "[School Mascot] Salad." Cafeteria staff affixed name cards to the serving line sneeze-guards in front of the vegetables for the day. For the fruit intervention, we purchased and delivered two ceramic bowls to the intervention school. The cafeteria staff filled the bowls and put them in different locations on the lunch line. For the milk intervention, cafeteria staff increased the amount of plain milk available by 10% and placed plain in front of chocolate milk in the coolers.
Experimental: Implement intervention for 15 weeks; In this school, we implemented the intervention on January 11, 2016, the day the study began.	Behavioral: Basic School Lunchroom Nudges; The intervention consists of three components: 1) serving fresh fruit in attractive bowls in at least two locations on the lunch line; 2) providing descriptive names for the vegetables; and 3) increasing amount of plain milk availability by 10% and placing it in front of the chocolate milk. For the vegetable naming intervention, researchers first developed, printed, and laminated descriptive vegetable name cards (2"x4"). Some examples included "Savory Collard Greens," and "[School Mascot] Salad." Cafeteria staff affixed name cards to the serving line sneeze-guards in front of the vegetables for the day. For the fruit intervention, we purchased and delivered two ceramic bowls to the intervention school. The cafeteria staff filled the bowls and put them in different locations on the lunch line. For the milk intervention, cafeteria staff increased the amount of plain milk available by 10% and placed plain in front of chocolate milk in the coolers.
Experimental: Implement intervention for 12 weeks; In this school, we collected baseline data for three weeks and then implemented the intervention for the remaining twelve weeks.	Behavioral: Basic School Lunchroom Nudges; The intervention consists of three components: 1) serving fresh fruit in attractive bowls in at least two locations on the lunch line; 2) providing descriptive names for the vegetables; and 3) increasing amount of plain milk availability by 10% and placing it in front of the chocolate milk. For the vegetable naming intervention, researchers first developed, printed, and laminated descriptive vegetable name cards (2"x4"). Some examples included "Savory Collard Greens," and "[School Mascot] Salad." Cafeteria staff affixed name cards to the serving line sneeze-guards in front of the vegetables for the day. For the fruit intervention, we purchased and delivered two ceramic bowls to the intervention school. The cafeteria staff filled the bowls and put them in different locations on the lunch line. For the milk intervention, cafeteria staff increased the amount of plain milk available by 10% and placed plain in front of chocolate milk in the coolers.
Experimental: Implement intervention for 9 weeks; In this school, we collected baseline data for six weeks and then implemented the intervention for the remaining nine weeks.	Behavioral: Basic School Lunchroom Nudges; The intervention consists of three components: 1) serving fresh fruit in attractive bowls in at least two locations on the lunch line; 2) providing descriptive names for the vegetables; and 3) increasing amount of plain milk availability by 10% and placing it in front of the chocolate milk. For the vegetable naming intervention, researchers first developed, printed, and laminated descriptive vegetable name cards (2"x4"). Some examples included "Savory Collard Greens," and "[School Mascot] Salad." Cafeteria staff affixed name cards to the serving line sneeze-guards in front of the vegetables for the day. For the fruit intervention, we purchased and delivered two ceramic bowls to the intervention school. The cafeteria staff filled the bowls and put them in different locations on the lunch line. For the milk intervention, cafeteria staff increased the amount of plain milk available by 10% and placed plain in front of chocolate milk in the coolers.
Experimental: Implement intervention for 6 weeks; In this school, we collected baseline data for nine weeks and then implemented the intervention for the remaining six weeks.	Behavioral: Basic School Lunchroom Nudges; The intervention consists of three components: 1) serving fresh fruit in attractive bowls in at least two locations on the lunch line; 2) providing descriptive names for the vegetables; and 3) increasing amount of plain milk availability by 10% and placing it in front of the chocolate milk. For the vegetable naming intervention, researchers first developed, printed, and laminated descriptive vegetable name cards (2"x4"). Some examples included "Savory Collard Greens," and "[School Mascot] Salad." Cafeteria staff affixed name cards to the serving line sneeze-guards in front of the vegetables for the day. For the fruit intervention, we purchased and delivered two ceramic bowls to the intervention school. The cafeteria staff filled the bowls and put them in different locations on the lunch line. For the milk intervention, cafeteria staff increased the amount of plain milk available by 10% and placed plain in front of chocolate milk in the coolers.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Percentage of students that took a serving of fruit	To determine whether or not a student took a serving of fruit, we will rely on tray waste records that we collected twice a week (repeated measures) for the duration of the study. Based on these records, if a student selected a serving of fruit, she will receive a value of one. Otherwise, she will receive a value of zero. We will use this binary measure to estimate the proportion of students who took a serving of fruit.	30 days
Percentage of students that took a serving of vegetables	To determine whether or not a student took a serving of vegetables, we will rely on tray waste records that we collected twice a week (repeated measures) for the duration of the study. Based on these records, if a student selected a serving of vegetables, she will receive a value of one. Otherwise, she will receive a value of zero. We will use this binary measure to estimate the proportion of students who took a serving of vegetables.	30 days
Percentage of students that took a carton of white milk (one pint)	To determine whether or not a student took a carton of white milk, we will rely on tray waste records that we collected twice a week (repeated measures) for the duration of the study. Based on these records, if a student selected a carton of white milk, she will receive a value of one. Otherwise, she will receive a value of zero. We will use this binary measure to estimate the proportion of students who took a carton of white milk.	30 days
Percentage of students that took a serving of flavored milk (one pint)	To determine whether or not a student took a carton of flavored milk, we will rely on tray waste records that we collected twice a week (repeated measures) for the duration of the study. Based on these records, if a student selected a carton of flavored milk, she will receive a value of one. Otherwise, she will receive a value of zero. We will use this binary measure to estimate the proportion of students who took a carton of flavored milk.	30 days
Percentage of students that took an entree	To determine whether or not a student took an entree, we will rely on tray waste records that we collected twice a week (repeated measures) for the duration of the study. Based on these records, if a student selected an entree, she will receive a value of one. Otherwise, she will receive a value of zero. We will use this binary measure to estimate the proportion of students who took an entree.	30 days
Fraction of fruit serving wasted by the student	The tray waste records our researchers collected twice a week (repeated measures) for the duration of the study indicate how much -- 0, one-quarter, one-half, three-quarters, all -- of the fruit serving a student threw away. We will use this measure to estimate the percentage of fruit servings students wasted.	30 days
Fraction of vegetable serving wasted by the student	The tray waste records our researchers collected twice a week (repeated measures) for the duration of the study indicate how much -- 0, one-quarter, one-half, three-quarters, all -- of the vegetable serving a student threw away. We will use this measure to estimate the percentage of vegetable servings students wasted.	30 days
Fraction of white milk wasted by the student	The tray waste records our researchers collected twice a week (repeated measures) for the duration of the study indicate how much -- 0, one-quarter, one-half, three-quarters, all -- of the serving of white milk (one pint) a student threw away. We will use this measure to estimate the percentage of white milk servings students wasted.	30 days
Fraction of flavored milk wasted by the student	The tray waste records our researchers collected twice a week (repeated measures) for the duration of the study indicate how much -- 0, one-quarter, one-half, three-quarters, all -- of the serving of flavored milk (one pint) a student threw away. We will use this measure to estimate the percentage of flavored milk servings students wasted.	30 days
Fraction of entree wasted by the student	The tray waste records our researchers collected twice a week (repeated measures) for the duration of the study indicate how much -- 0, one-quarter, one-half, three-quarters, all -- of the entree serving a student threw away. We will use this measure to estimate the percentage of entree servings students wasted.	30 days
Proportion of prepared fruit servings that students took	Cafeterias in the study provided us with food production records that provide daily records of the number of fruit servings cafeteria staff prepared and the number of fruit servings students took. To calculate the proportion of prepared fruit servings taken by students we will divide the number of servings students took by the total number of servings prepared.	75 days
Proportion of prepared vegetable servings that students took	Cafeterias in the study provided us with food production records that provide daily records of the number of vegetable servings cafeteria staff prepared and the number of vegetable servings students took. To calculate the proportion of prepared vegetable servings taken by students we will divide the number of servings students took by the total number of servings prepared.	75 days
Proportion of prepared entree servings that students took	Cafeterias in the study provided us with food production records that provide daily records of the number of entree servings cafeteria staff prepared and the number of entree servings students took. To calculate the proportion of prepared entree servings taken by students we will divide the number of servings students took by the total number of servings prepared.	75 days
Proportion of white milk cartons taken	Cafeterias in the study provided us with food production records that provide daily records of the number of white milk cartons made available and the number of white milk cartons students took. To calculate the proportion of white milk cartons taken by students we will divide the number of servings students took by the total number of servings prepared.	75 days
Proportion of flavored milk cartons taken	Cafeterias in the study provided us with food production records that provide daily records of the number of flavored milk cartons made available and the number of flavored milk cartons students took. To calculate the proportion of flavored milk cartons taken by students we will divide the number of servings students took by the total number of servings prepared.	75 days
Percentage of total students that took a fruit serving	Cafeterias in the study provided us with food production records that provide daily records of the number of fruit servings prepared and the total number of students who received a school lunch. To calculate the percentage of students who took a fruit serving we will divide the number of servings students took by the total number of students who received lunch.	75 days
Percentage of total students that took a vegetable serving	Cafeterias in the study provided us with food production records that provide daily records of the number of vegetable servings prepared and the total number of students who received a school lunch. To calculate the percentage of students who took a vegetable serving we will divide the number of servings students took by the total number of students who received lunch.	75 days
Percentage of total students that took an entree serving	Cafeterias in the study provided us with food production records that provide daily records of the number of entree servings prepared and the total number of students who received a school lunch. To calculate the percentage of students who took an entree serving we will divide the number of servings students took by the total number of students who received lunch.	75 days
Percentage of total students that took a carton of white milk	Cafeterias in the study provided us with food production records that provide daily records of the number of white milk cartons made available the total number of students who received a school lunch. To calculate the percentage of students who took a carton of white milk we will divide the number of servings students took by the total number of students who received lunch.	75 days
Percentage of total students that took a carton of flavored milk	Cafeterias in the study provided us with food production records that provide daily records of the number of flavored milk cartons made available the total number of students who received a school lunch. To calculate the percentage of students who took a carton of flavored milk we will divide the number of servings students took by the total number of students who received lunch.	75 days
Percentage of fidelity assessments that returned positive marks	Throughout the study, researchers conducted fidelity assessments to make sure the schools maintained the interventions. These process control measures are the ratio of successful fidelity checks to the total number. Researchers conducted these assessments during each day they collected tray waste measures and during a randomly selected non-tray waste collection date once a month.	36 days

Collaborators and Investigators

• Sponsor: Ohio State University
• Collaborators: Cornell University; Columbus City School District
• Investigators: Principal Investigator: Andrew S Hanks, PhD, Ohio State University; Study Director: Haleigh Gaines, MS,RD, Ohio State University

Study record dates

Study Major Dates

• Study Start (Actual): January 11, 2016
• Primary Completion (Actual): April 27, 2016
• Study Completion (Actual): April 29, 2016

Study Registration Dates

• First Submitted: October 29, 2019
• First Submitted That Met QC Criteria: November 6, 2019
• First Posted (Actual): November 7, 2019

Study Record Updates

• Last Update Posted (Actual): November 7, 2019
• Last Update Submitted That Met QC Criteria: November 6, 2019
• Last Verified: November 1, 2019

More Information

Terms related to this study

• Keywords: Child nutrition; National School Lunch Program; Behavioral nudges; Food environment; Choice architecture; Lasting behavior change
• Other Study ID Numbers: 2016B0004; 2019B0480 (Other Identifier: Ohio State University); 62140-10509 (Other Grant/Funding Number: Cornell University)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Plan Description: We agree to share the study data with researchers who request access. While we will not publicly post the data, we will provide them through an electronic link. None of the data contain unique identifiers.
• IPD Sharing Time Frame: Data are available as of October 29, 2019 and will be available until December 31, 2020.
• IPD Sharing Access Criteria: Those interested in acquiring the data can contact Dr. Andrew Hanks at hanks.46@osu.edu.
• IPD Sharing Supporting Information Type: STUDY_PROTOCOL; ANALYTIC_CODE

Drug and device information, study documents

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