Table 2. Time-Based Event Rates (%) of SARS-CoV-2 Infections in the Matched Cohort.
| Follow-up, d |
Vaccinated | Unvaccinated | |||||
|---|---|---|---|---|---|---|---|
| Uncensored raw rate a (events/women) | Hazard rate accounting for censoring b | Uncensored raw rate a (events/women) | Hazard rate accounting for censoring b | Absolute difference of hazard rate (95% CI) | Adjusted hazard ratio c (95% CI) | P value | |
| Primary analysis | |||||||
| ≤10 |
0.93 (70/7530) | 0.93 | 0.97 (73/7530) | 0.97 | 0.04 (–0.18 to 0.25) | 0.96 (0.69-1.33) | .79 |
| 11-27 |
0.51 (38/7387) | 0.60 | 1.12 (83/7387) | 1.34 | 0.74 (0.48 to 1.00) | 0.46 (0.31-0.67) | <.001 |
| ≥28d |
0.21 (10/4788) | 0.33 | 0.96 (46/4788) | 1.64 | 1.31 (0.89 to 1.74) | 0.22 (0.11-0.43) | <.001 |
| Post hoc analysis | Women, No. | Cumulative incidence c | Women, No. | Cumulative incidence e | Absolute difference of cumulative incidence (95% CI) | ||
| 10 | 7403 | 0.93 | 7403 | 0.97 | 0.04 (–0.18 to 0.25) | ||
| 27 | 4903 | 1.53 | 4903 | 2.30 | 0.77 (0.54 to 1.12) | ||
| 28 | 4788 | 1.55 | 4788 | 2.34 | 0.80 (0.47 to 1.13) | ||
| 35 | 4023 | 1.68 | 4023 | 2.72 | 1.04 (0.67 to 1.41) | ||
| 42 | 3376 | 1.68 | 3376 | 2.96 | 1.28 (0.87 to 1.65) | ||
| 49 | 2327 | 1.72 | 2327 | 3.09 | 1.37 (0.96 to 1.75) | ||
| 56 | 1748 | 1.77 | 1748 | 3.48 | 1.71 (1.26 to 2.17) | ||
| 63 | 1295 | 1.77 | 1295 | 3.75 | 1.97 (1.46 to 2.47) | ||
| 70 | 955 | 1.85 | 955 | 3.90 | 2.05 (1.53 to 2.57) | ||
Raw rate (not accounting for censoring) was calculated as the number of events during the period divided by the number of women at risk at the beginning of the period.
Hazard rate accounting for censoring for each period was calculated as the survival at the beginning of the period minus survival at the end of it divided by the survival at the beginning of the period. Survival (1 − cumulative incidence) was estimated with the Kaplan-Meier method to account for censoring. The absolute difference of hazards between vaccinated and unvaccinated is reported alongside CIs, which were calculated with the bootstrap percentile method with 500 samples of matched pairs.
Hazard ratio during 28 days or more of follow-up was the primary outcome, calculated with a Cox time-varying hazard model adjusted for population subgroup, maternal age, gestational age, influenza vaccine, number of prior children, infertility, and obesity.
The maximum time of follow-up was 110 days and the median was 37 days (interquartile range, 21-54 days) for both groups.
Cumulative incidence was calculated as 1 minus survival probability at specific days during the follow-up until the maximal point at which at least 10% of the matched cohort remained under follow-up. The difference of cumulative incidence rates between vaccinated and unvaccinated is reported alongside CIs, which were calculated with the bootstrap percentile method with 500 samples of matched pairs.