Table 1.

Chatbot characteristics.

	Proprietary name	Country	Vaccines	Use cases	Chatbot type	Development platform	Deployment platform	Evaluation methodology	Outcome variables	Theoretical underpinning
Altay et al [19]	None	France	COVID-19	Information provision	Rules-based	Not specified	Custom-built web page	Randomized control trial (n=701)	COVID-19 vaccine attitudes; COVID-19 vaccine intent	None stated
Hong et al [20]	None	South Korea	Childhood vaccines as per national immunization schedule	Information provision; vaccine scheduling; appointment reminders; misinformation debunking; financial incentives	Natural language	Google Dialog-flow	Kakao Plus Friend	Quasi-experiment (n=65)	Vaccination information; Vaccination motivation; Self-efficacy; Vaccination behavioral intention	Information-motivation behavioral skills model
Chalaguine and Hunter [22]	None	United Kingdom (inferred)	COVID-19	Vaccine counseling or persuasion; information provision	Natural language	Javascript and Python	Flask web server	Experiment (n=300)	COVID-19 vaccine intent	None stated
Kobayashi et al [21]	Corowa-kun	Japan	COVID-19	Information provision	Hybrid rules-based or natural language	Not specified	LINE	Cross-sectional survey (n=10,192)	COVID-19 vaccine intent	None stated
Amith et al [16,17]	None	United States	HPVa	Vaccine counseling or persuasion; information provision	Simulated conversational agent (“Wizard of Oz”)	Apple SDKb	iPad app	Pre- and postuse surveys (2019: n=18; 2020: n=24)	Vaccine hesitancy; Perceived chatbot usability	Health Belief Model
Tsai et al [18]	None	United States	HPV	Information provision	Simulated conversational agent (“Wizard of Oz”)	tawk.to	Website	Experiment (n=142)	Satisfaction with chatbot; Perceived chatbot utility; HPV vaccine intent	Agency effect

^aHPV: human papillomavirus.

^bSDK: Software Development Kit.