Abstract
OBJECTIVE: The COVID-19 vaccination is a key strategy to control the pandemic; however, complex factors, including health awareness and social cognition, influence public intention to vaccinate. The Health Belief Model (HBM) provides a theoretical framework for understanding vaccination behavior, but how Vaccination awareness (VA) dynamically moderates the relationship between HBM domains and vaccination intentions remains unclear. This study aims to compare the characteristics of different VA classifications and explore the key factors influencing their future COVID-19 vaccination intentions based on the HBM. METHODS: Using three-wave longitudinal cohort data from 500 adults in mainland China, participants were divided into four groups based on VA states: persistent awareness (Group 1), early-only awareness (Group 2), late-emerging awareness (Group 3), and persistently unaware (Group 4). ANOVA, chi-square tests, and binary logistic regression were used to analyze the relationships between HBM constructs, social cues, and vaccination intentions. RESULTS: Group 1 (31.40%) had the highest education level, the COVID-19 vaccination rate (48.35%), and future vaccination intention (36.98%), with significantly higher HBM scores (perceived severity, self-efficacy, benefits) than other groups (all p < 0.001). Group 4 (27.00%) had the highest proportion of unvaccinated individuals (40.31%) and the highest perceived barriers score (M = 2.39 ± 0.73). Logistic regression showed that Group 1's future COVID-19 vaccination intention was driven by medical expert advice (OR = 3.08, 95% CI: 1.47-6.45) and self-efficacy (OR = 3.68, 95% CI: 1.53-8.87), while Group 4 relied on government advice (OR = 1.56, 95% CI: 1.02-2.40), perceived severity (OR = 1.97, 95% CI: 1.12-3.46), and self-efficacy (OR = 3.68, 95% CI: 1.77-5.84). Trust in social media significantly reduced Group 3's vaccination intention (OR = 0.08, 95% CI: 0.02-0.35). CONCLUSIONS: Vaccination awareness states influence vaccination decisions by moderating HBM domains and social cues. Targeted interventions should address group-specific differences, such as enhancing self-efficacy (for Group 1, Group 2, and Group 4), correcting social media misinformation ( for Group 3), and strengthening government communication (for Group 4).