A Conversational AI System to Enhance Knowledge About Vaccines and Promote Vaccination

A Conversational AI System to Enhance Knowledge About Vaccines and Promote Vaccination: A Randomized Controlled Trial

This study is designed to contribute to the growing body of evidence on the efficacy of digital health interventions. By rigorously evaluating the impact of the AI system on vaccine knowledge and acceptance, the study aims to inform future health communication strategies and policies. It stands as a potential model for integrating AI into public health initiatives, showcasing how technology can be harnessed to improve health outcomes on a large scale.

Study Overview

• Status: Completed
• Conditions: Vaccination Hesitancy
• Intervention / Treatment: Other: Printed educational Brochure; Device: Conversational AI system

• Study Type: Interventional
• Enrollment (Actual): 430
• Phase: Not Applicable

Contacts and Locations

Study Locations

• China
  • Guangdong
    • Guangzhou, Guangdong, China, 510000
      • Zhongshan Ophthalmic Center

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 60 years and older (Adult, Older Adult)
• Accepts Healthy Volunteers: Yes

Description

Inclusion Criteria

• Individuals aged 18 or above
• Owning a smartphone/tablet capable of using the study app
• Willing to provide informed consent.

Exclusion Criteria

• Individuals with ocular or hearing disorders that preclude the use of mobile phones or reading.
• Individuals with mental disorders that preclude the use of mobile phones or reading.

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Health Services Research
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Triple

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: Conversational AI system	Device: Conversational AI system; The conversational AI system based on LLM processes user queries with a sophisticated generation, decision, action, and safety framework to provide optimal feedback. The system has several key features. Firstly, it utilizes retrieval-augmented in-context learning to enhance responses generated by the LLM, sourcing information from trustworthy websites. Secondly, it includes Guardrail module that addresses potential harmful content in the LLM-generated responses by clarifying and validating the content before delivery. Furthermore, the AI system is equipped with Self-checking memory module that retains essential clinical characteristics across multi-turn dialogues, ensuring the continuity and consistency of its interactions with users. The educational content about vaccination was derived from the official website of World Health Organization (https://www.who.int/news-room/questions-and-answers/item/vaccines-and-immunization-what-is-vaccination).
Active Comparator: Educational brochure	Other: Printed educational Brochure; A printed brochure whose content about vaccination was derived from the official website of World Health Organization (https://www.who.int/news-room/questions-and-answers/item/vaccines-and-immunization-what-is-vaccination).

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Acceptance of vaccines	Vaccine acceptance was evaluated based on responses to the question, 'Based on your actual situation, please choose the extent of your willingness to get vaccinated.' The response options were: Absolutely refuse, Refuse but uncertain about the future, Have not considered it yet, Accept but still considering, and Completely willing. Vaccine acceptance was considered if the participant selected 'Completely willing.	Day 14

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Hesitancy of vaccination	The Vaccine Hesitancy Scale was developed based on the "3Cs" model (confidence, complacency, and convenience) by the SAGE Working Group on Vaccine Hesitancy, encompassing three aspects: confidence, complacency, and convenience. It utilizes a Likert 5-point scale (Strongly Disagree, Disagree, Neutral, Agree, Strongly Agree) for scoring. For the complacency dimension, aspect, scoring is reversed with "Strongly Agree" to "Strongly Disagree" rated as 5 to 1 points, respectively. In contrast, other items follow the standard scoring, where "Strongly Agree" to "Strongly Disagree" are scored from 1 to 5 points. Confidence is measured by perceived vaccine safety and efficacy. Complacency is assessed through perceived necessity of the vaccine and the severity of the preventable diseases. Convenience is evaluated by the perceived ease of transportation and appointment scheduling.	Day 14

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
Feelings of coercion	To assess feelings of coercion related to vaccination, participants were asked a single question: 'When deciding whether to receive the vaccines, do you feel forced to either take or not take the vaccine?' This question utilizes a 5-point Likert scale (very low, low, neutral, high, very high) with lower scores indicating lower feelings of perceived coercion.	Day 14

Collaborators and Investigators

• Sponsor: Sun Yat-sen University

Study record dates

Study Major Dates

• Study Start (Actual): May 1, 2023
• Primary Completion (Actual): August 31, 2023
• Study Completion (Actual): August 31, 2023

Study Registration Dates

• First Submitted: January 16, 2023
• First Submitted That Met QC Criteria: January 16, 2023
• First Posted (Actual): January 18, 2023

Study Record Updates

• Last Update Posted (Estimated): January 19, 2024
• Last Update Submitted That Met QC Criteria: January 17, 2024
• Last Verified: January 1, 2024

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Coronavirus Infections; Coronaviridae Infections; Nidovirales Infections; RNA Virus Infections; Virus Diseases; Infections; Respiratory Tract Infections; Respiratory Tract Diseases; Pneumonia, Viral; Pneumonia; Lung Diseases; COVID-19
• Other Study ID Numbers: 2021KYPJ182

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