Vaccine uptake in the context of mandate announcement and removal: Evidence from Europe and North America

BACKGROUND: The COVID-19 pandemic drove the implementation of vaccine mandates to protect spaces and/or increase vaccine uptake, though their design, timing, and scope varied across jurisdictions. This study examines the associations of vaccine mandate announcements and removals with COVID-19 vaccine uptake in France, Italy, and California (USA). METHODS: We employed interrupted time series (ITS) analysis using aggregated data from Our World in Data and the Oxford COVID-19 Government Response Tracker. The primary outcomes were the weekly number of first and booster doses per 100,000 people. We estimated changes in the rate and trend of vaccine uptake following policy announcements and removals, adjusting for COVID-19 cases, deaths, and temporal autocorrelation to account for other influencing factors. FINDINGS: Initial mandate announcements were associated with immediate increases in vaccine uptake: 113% (95% CI: 63 to 178%) in Italy, 195% (95% CI: 113 to 308%) in France, and 32% (95% CI: 16 to 51%) in California. However, these associations attenuated in more fully adjusted models, especially in France, where the initial association was no longer statistically significant after controlling for epidemiological conditions and temporal autocorrelation. Mandate removals were more consistently associated with declines in booster uptake in Italy and France, including declines of 62% (95% CI: -74 to -45%) and 70% (95% CI: -79 to -57%), respectively, while partial removal in California coincided with increased booster uptake. CONCLUSION: Vaccine mandate announcements were associated with short-term increases in vaccine uptake during key periods of the pandemic. Their removal was often associated with a slowing in uptake, highlighting the need for coordinated communication strategies to sustain coverage. Future policy responses should consider timing, mandate design, and transition planning in contexts where such policies are used.