A field experiment on information provision for promoting citizens’ health checkup behavior: Postal mail or SMS?

Last registered on January 12, 2024
View Trial History

Pre-Trial

Trial Information

General Information

Title
A field experiment on information provision for promoting citizens’ health checkup behavior: Postal mail or SMS?
RCT ID
AEARCTR-0012828
Initial registration date
January 12, 2024

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
January 12, 2024, 3:40 PM EST

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

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

Request Information

Primary Investigator

Name
Shusaku Sasaki
Affiliation
Osaka University
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Midori Noguchi
PI Affiliation
Osaka University
Contact Investigator
PI Name
Asuka Oyama
PI Affiliation
Osaka University
Contact Investigator
PI Name
Fumio Ohtake
PI Affiliation
Osaka University
Contact Investigator

Additional Trial Information

Status
On going
Start date
2023-11-02
End date
2026-03-31
Keywords
Behavior, Health
Additional Keywords
Health behavior, Health checkup, Health screening, Nudge, Health economics Behavioral economics, Experimental economics, Field experiment, Japan
JEL code(s)
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
In 2008, the Japanese government launched a new health checkup system, the Specified Health Examination. The purpose of this checkup is to screen pre-diabetic patients who are at risk of developing cardiovascular diseases and diabetes mellitus, and to provide early intervention for receiving medical treatment and improving lifestyle. The uptake rate in FY2020 was 53.4%, compared to the government’s target of 70%. The uptake rate for National Health Insurance subscribers, which includes self-employed people, was only 33.7%.

In this study, we collaborate with a Japanese local government and conduct two experiments focusing on the channels (postal mail and SMS) used to send reminder messages to encourage National Health Insurance subscribers to receive health checkups. In one experiment, we measure the effects of two reminder messages based on behavioral economics and nudge through the postal mail channel. In the other experiment, we compare the effect of sending a reminder message through the postal mail channel with that of sending it through the SMS channel.

The subjects of this experiment (around 40,000 citizens) are the National Health Insurance subscribers in the age group eligible for the Specified Health Examination in FY2023. In addition, they have not received the checkups by the time the postcards are mailed. First, we divide 40,000 citizens into two groups: the subjects of Experiment 1 (around 28,000 citizens) and the subjects of Experiment 2 (around 11,000 citizens). The former are citizens whose mobile phone numbers are not known by the local government and to whom SMS cannot be sent. The latter are citizens whose mobile phone numbers are known to the local government and to whom SMS can be sent. Second, in Experiment 1, we randomly assign around 28,000 citizens to the following three groups at individual level:

- Control group 1A: We mail a postcard explaining the necessity of the specific health examination and how to make an appointment.
- Treatment group 1B: We add the message "Do you still have your free voucher for the specific health examination?" to the control group's postcard and then introduce a simple way to reissue the voucher.
- Treatment group 1C: We add the message "Your voucher for health checkup is about to expire" to the control group's postcard and then call for early reservations.

In Experiment 2, we randomly assign around 11,000 citizens to the following three groups at individual level:

- Control group 2A: No reminder
- Treatment group 2B: We send a reminder message through the postal mail channel.
- Treatment group 2C: We send a reminder message through the SMS channel.

We will send the postcards in January 2024. The outcome variables are the presence or absence and timing of the health checkup from the first mailing time to the end of March 2024.
External Link(s)

Registration Citation

Citation
Noguchi, Midori et al. 2024. "A field experiment on information provision for promoting citizens’ health checkup behavior: Postal mail or SMS?." AEA RCT Registry. January 12. https://doi.org/10.1257/rct.12828-1.0
Sponsors & Partners
Experimental Details

Interventions

Intervention(s)
First, we divide 40,000 citizens into two groups: the subjects of Experiment 1 (around 28,000 citizens) and the subjects of Experiment 2 (around 11,000 citizens) and respectively perform stratified randomization. The former are citizens whose mobile phone numbers are not known by the local government and to whom SMS cannot be sent. The latter are citizens whose mobile phone numbers are known to the local government and to whom SMS can be sent. Second, in Experiment 1, we randomly assign around 28,000 citizens to the following three groups at individual level:

- Control group 1A: We mail a postcard explaining the necessity of the specific health examination and how to make an appointment.
- Treatment group 1B: We add the message "Do you still have your free voucher for the specific health examination?" to the control group's postcard and then introduce a simple way to reissue the voucher.
- Treatment group 1C: We add the message "Your voucher for health checkup is about to expire" to the control group's postcard and then call for early reservations.

In Experiment 2, we randomly assign around 11,000 citizens to the following three groups at individual level:

- Control group 2A: No reminder
- Treatment group 2B: We send a reminder message through the postal mail channel.
- Treatment group 2C: We send a reminder message through the SMS channel.

We will send the postcards in January 2024.
Intervention Start Date
2024-01-16
Intervention End Date
2024-03-31

Primary Outcomes

Primary Outcomes (end points)
Whether a subject receives the specific health examination or not
Primary Outcomes (explanation)
The outcome variables will be collected by the local government as administrative data and then shared with the researchers under a nondisclosure agreement.

Secondary Outcomes

Secondary Outcomes (end points)
When a subject receives the specific health examination
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Respectively for two experiments, we perform stratified randomization in each of the fourteen strata. We construct the seven strata based on the subject’s sex, age and probability of receiving the specific health examination. We predict the individual probability, using the information on their past uptake behavior and health examination results.

We will exclude from the analysis those who receive the checkup between the timings of data sampling and the first mailing or those who contact the local government and make an application to opt-out from the analysis of this experiment.

We use the age information provided in the administrative data and perform a balance check for each stratum. We employ a 5% significance level for this test.

Our primary outcome variable is whether or not a subject receives the specific health examination from the first mailing time to the end of March 2024. First, we evaluate each treatment effect using a full sample and t-tests. We employ a 5% significance level for this test as well.

Second, we evaluate each treatment effect using a full sample and regression analysis. We also compare the treatment effects with each other. In the regression analysis, we control for sex, age and dummy variables for each stratum.

Third, we use regression analysis and evaluate the treatment effects by strata to examine their heterogeneity. We similarly compare the effects with each other.
Experimental Design Details
Not available
Randomization Method
First, we divide 40,000 citizens into two groups: the subjects of Experiment 1 (around 28,000 citizens) and the subjects of Experiment 2 (around 11,000 citizens) and respectively perform stratified randomization. The former are citizens whose mobile phone numbers are not known by the local government and to whom SMS cannot be sent. The latter are citizens whose mobile phone numbers are known to the local government and to whom SMS can be sent.

We construct the seven strata based on the subject’s sex, age and probability of receiving the specific health examination. We predict the individual probability, using the information on their past uptake behavior and health examination results.
Randomization Unit
Individuals.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
N/A
Sample size: planned number of observations
Experiment 1: around 28,000 citizens Experiment 2: around 11,000 citizens
Sample size (or number of clusters) by treatment arms
Experiment 1: around 9,500 citizens for each group
Experiment 2: around 3,800 citizens for each group
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
N/A
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
Center for Infectious Disease Education and Research, Osaka University IRB
IRB Approval Date
2023-10-25
IRB Approval Number
2023CRER1025
Analysis Plan