BACKGROUND: Rogers' diffusion of innovation theory suggests innovations are adopted in stages by different groups (innovators/early adopters, early majority, late majority and late adopters). In healthcare, this could mean that there is the potential to worsen health disparities as later groups tend to also face more social and structural barriers. Determining the unique sociodemographic characteristics, beliefs and attitudes of those in each diffusion category could be useful for theorising how to reach later groups more effectively. METHODS: Using a cross-sectional survey among Canadian adults in late 2021, we assigned respondents to diffusion groups based on when they received their first dose, relative to others within their age group in accordance with Rogers' model (ie, cut points: 16%, 50%, 84% with 100% being all those vaccinated within the age group). Participants answered questions about their COVID-19 vaccinations and questions related to their motivations, beliefs, values and attitudes towards COVID-19. A multinomial logistic regression model assessed the likelihood of participants being associated with each diffusion category (with the significance level set at p<
0.05). RESULTS: The final sample included 2131 respondents. Late adopters were significantly more likely to identify as non-white, live in rural locations and receive vaccinations at pharmacies. Innovators and early adopters were significantly more likely to get vaccinated in settings other than pharmacies or community centres. CONCLUSION: A diffusion group-based analysis brought insight into how vaccination strategies could be tailored to reach each diffusion group sooner, particularly late adopters who encounter more barriers.