Table 2.
Vaccine effectiveness measures
| Delta period | Omicron period | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Hospitalization | Significant disease | Severe disease | Hospitalization | Significant disease | Severe disease | |||||||
| HR | 1-HR (95% CI) | HR | 1-HR (95% CI) | HR | 1-HR (95% CI) | HR | 1-HR (95% CI) | HR | 1-HR (95% CI) | HR | 1-HR (95% CI) | |
| 3rd vs. 2nd vaccine group | 0.08 (0.04–0.17) | 92% (83–96) | 0.15ab (0.01–1.83) | 0 vs. 4 | 0.17ab (0.01–2.40) | 0 vs. 3 | 0.52 (0.43–0.63) | 48% (37–57) | 0.08 (0.01–0.74) | 92% (26–99) | 0.41ab (0.02–6.68) | 0 vs. 1 |
| 2nd dose vs. unvaccinated group | 0.39 (0.31–0.49) | 61% (51–69) | 0.03 (0.01–0.08) | 97% (92–99) | 0.04 (0.01–0.14) | 96% (86–99) | 1.12 (0.92–1.36) | [−12%] ([−36]−8) | 0.49 (0.16–1.47) | 51%([−47]−84) | 0.17 (0.02–1.47) | 83% ([−47]−98) |
| 3rd dose vs. unvaccinated group | 0.03 (0.01–0.05) | 97% (95–99) | 0.01a (0.00–0.07) | 99% (93–100) 0 vs. 108 | 0.02 a (0.00–0.11) | 99% (89–100) 0 vs. 64 | 0.57 (0.47–0.69) | 43% (31–53) | 0.03 (0.00–0.28) | 97% (72–100) | 0.06a (0.01–0.57) | 94% (43–99) 0 vs. 5 |
HR of COVID-19 hospitalization degree for vaccinated group vs. unvaccinated group and between the vaccinated groups, in two study periods [Delta period- 1 August 2021–1 December 2021; Omicron period- 15 December 2021–22 March 2022]. The study period populations were 82,659 and 33,303 in each of the two periods, respectively. In each study period, 28,213 and 714 individuals were first included in the 2-dose group and then re-recruited to the 3-dose group.
HR calculated using Cox proportional-hazards regression model with time-dependent covariates controlling for maternal age, parity and days of follow-up.
Outcomes definitions: hospitalization with SARS-CoV-2 infection; significant disease (e.g., COVID-19-related pneumonia associated with COVID-19 justifying hospitalization), and severe disease (e.g., resting respiratory rate >30 breaths per minute, O2 saturation on room air <94%, etc.).
aA single case was imputed to allow estimations of HR. Estimation is based on the mean of 1000 simulations.
bEstimates (1-HR) were only calculated for cells with 5 events or more, otherwise, raw counts are reported.