Beliefs about and confidence in new medical treatments: Lessons from COVID-19 vaccines

How to cite

Nakabayashi, Masaki; Kawata, Keisuke; Fukai, Taiyo. 2022. "Beliefs about and confidence in new medical treatments: Lessons from COVID-19 vaccines." AEA RCT Registry. AEARCTR-0010145. https://doi.org/10.1257/rct.10145-1.1

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Masaki Nakabayashi* , Keisuke Kawata , Taiyo Fukai

* Primary investigator

Registry ID
AEARCTR-0010145
DOI
https://doi.org/10.1257/rct.10145-1.1
First registered
Last updated
Status
Completed
Trial dates
2022-11-09 – 2022-11-24
Countries
Japan ()
Registry record
http://www.socialscienceregistry.org/trials/10145

Abstract

1. Background and study aims.
Public health depends on both the technological developments of medical treatments and the public's willingness to take up these treatments. The public’s willingness to take up novel treatments depends on the public's beliefs about the costs and benefits of the treatments. By tracing updates on public beliefs about COVID-19 vaccines before and after they began to be administered to the general public in Japan, we investigate the relationships between the evolution of public beliefs about COVID-19 vaccines and the public willingness to take up the vaccines.

2. Who can participate?
We recruit a non-probability sample of 15,000 respondents in Japan through a survey company, Rakuten Insight, Ltd.

3. What does the study involve?
1) We ask whether respondents have taken COVID-19 vaccines and, if so, whether they experienced side effects.
2) We ask whether the respondents have taken a second and third dose of COVID-19 vaccines and whether they want to take a fourth dose of COVID-19 vaccine.
3) We ask the respondents to choose the preferred hypothetical conditions for vaccination between the two, which are fully randomly generated by our conjoint experimental design. We allow the respondents to answer to deny either condition and not to take up COVID-19 vaccines.
4) We ask the respondents questions about their demographic and socioeconomic backgrounds.

4. What are the possible benefits and risks of participating?
As a benefit, respondents will receive a certain number of points to be used for shopping. Since the survey is administered through the internet, respondents will face no substantial risk.

5. Where is the study run from?
The survey will be conducted through the internet.

6. When is the study starting and how long is it expected to run for?
We plan to begin the survey on November 1, 2022 and continue until we obtain responses from 15,000 respondents. The duration is expected to be two weeks.

7. Who is funding the study?
The Japan Society for the Promotion of Science funds the study (SPS Grant-in-Aid KAKENHI JP18H03630).

Intervention

We investigate whether belief updates about the costs and benefits of medical treatments affect confidence in medical treatments, taking an example from vaccinations for COVID-19. Testable hypotheses are:
1) Unexpected side effects, which are part of the costs, affect confidence in vaccination.
2) Changes in the financial and time costs of vaccination affect willingness to take up vaccination.

Our fully randomized conjoint design generates various conditions for taking up COVID-19 vaccines. Since attributes of the hypothetical conditions are fully randomized, we can identify a change in attributes and a change in willingness to take up COVID-19 vaccines as a causal effect. Additionally, we investigate whether the causal channel is affected by past experience of "COVID arm" symptoms.

Experimental design

This study is the final stage of a longitudinal interventional study that began in the first wave in February 2021, when vaccines for COVID-19 were not available to the general public in Japan, followed by the second wave in March 2022, when COVID-19 vaccines had been available to the general public.

We have asked whether respondents had experienced the delayed localized hypersensitivity reactions to COVID-19 vaccines, which are referred to as "COVID arm" and were mostly unexpected to most of the public, conditional on vaccines take up, as the vaccines were novel. We use the "COVID arm" symptoms as our first treatment that could have updated the beliefs about the costs of medical treatment.

1) In this wave, we ask respondents whether they took a third dose of COVID-19 vaccine and investigate whether the probability of taking the third dose was associated with having experienced the "COVID arm" symptoms after taking up first or second dose of a COVID-19 vaccine.
2) Additionally, we design a fully randomized conjoint experiment to generate two hypothetical conditions of vaccination against COVID-19, where we allow respondents to accept or decline vaccines, and we ask respondents under which conditions they might more likely take up COVID-19 vaccines.

Design details

Randomization method
Randomization will be done in the office by a computer.
Randomization unit
Individual
Clusters
15,000 individuals.
Observations
Since we assign 5 conjoint experimental tasks to each indvidual, we plan to obtain 15,000X5=450,000 observations.
Arms
Since our treatment is a fully randomized conjoint design, all 15,000 respondents are included in the treatment arms.
Primary outcomes
Our primary outcome measures are self-reported subjective cost of/confidence in vaccination. In addition, we also measure self-reported history of taking up COVID-19 vaccines. Additionally, we measure any impact of side effects on beliefs about vaccines, and through them, confidence in vaccination, and how long the change in confidence in vaccination sustains. Our previous study demonstrated that experiencing symptoms of "COVID arm" after a first or second dose of COVID-19 vaccine adversely affected confidence in vaccination and willingness to take a third dose of COVID-19 vaccines. If the causal effect from belief updates to confidence in vaccination is long-lived, the adverse effects are expected to also be observed in this round. If the adverse effects of "COVID arm" symptoms by March 2022 are not observed in November 2022, the effects of belief updates are considered to have been short-lived, and the risk of once unexpected side effects are considered to have been factored in.
Secondary outcomes
Furthermore, we generate hypothetical vaccination conditions under which the costs of taking up vaccines vary and ask respondents which conditions are preferred. Additionally, if the adverse effects of belief updates about “COVID arm” are long-lived and related to other dimensions of costs for vaccination, the results of our conjoint experiment will be associated with experiences of “COVID-19 arm” symptoms.

Cited by

No citation data available for this registration yet.

Keywords