Waning effectiveness of BNT162b2 and ChAdOx1 COVID-19 vaccines over six months since second dose: a cohort study using linked electronic health records

Abstract

Background The rate at which COVID-19 vaccine effectiveness wanes over time is crucial for vaccination policies, but is incompletely understood with conflicting results from different studies. Methods This cohort study, using the OpenSAFELY-TPP database and approved by NHS England, included individuals without prior SARS-CoV-2 infection assigned to vaccines priority groups 2-12 defined by the UK Joint Committee on Vaccination and Immunisation. We compared individuals who had received two doses of BNT162b2 or ChAdOx1 with unvaccinated individuals during six 4-week comparison periods, separately for subgroups aged 65+ years; 16-64 years and clinically vulnerable; 40-64 years and 18-39 years. We used Cox regression, stratified by first dose eligibility and geographical region and controlled for calendar time, to estimate adjusted hazard ratios (aHRs) comparing vaccinated with unvaccinated individuals, and quantified waning vaccine effectiveness as ratios of aHRs per-4-week period. The outcomes were COVID-19 hospitalisation, COVID-19 death, positive SARS-CoV-2 test, and non-COVID-19 death. Findings The BNT162b2, ChAdOx1 and unvaccinated groups comprised 1,773,970, 2,961,011 and 2,433,988 individuals, respectively. Waning of vaccine effectiveness was similar across outcomes and vaccine brands: e.g. in the 65+ years subgroup ratios of aHRs versus unvaccinated for COVID-19 hospitalisation, COVID-19 death and positive SARS-CoV-2 test ranged from 1.23 (95% CI 1.15-1.32) to 1.27 (1.20-1.34) for BNT162b2 and 1.16 (0.98-1.37) to 1.20 (1.14-1.27) for ChAdOx1. Despite waning, rates of COVID-19 hospitalisation and COVID-19 death were substantially lower among vaccinated individuals compared to unvaccinated individuals up to 26 weeks after second dose, with estimated aHRs <0.20 (>80% vaccine effectiveness) for BNT162b2, and <0.26 (>74%) for ChAdOx1. By weeks 23-26, rates of SARS-CoV-2 infection in fully vaccinated individuals were similar to or higher than those in unvaccinated individuals: aHRs ranged from 0.85 (0.78-0.92) to 1.53 (1.07-2.18) for BNT162b2, and 1.21 (1.13-1.30) to 1.99 (1.94-2.05) for ChAdOx1. Interpretation The rate at which estimated vaccine effectiveness waned was strikingly consistent for COVID-19 hospitalisation, COVID-19 death and positive SARS-CoV-2 test, and similar across subgroups defined by age and clinical vulnerability. If sustained to outcomes of infection with the Omicron variant and to booster vaccination, these findings will facilitate scheduling of booster vaccination doses.

The Vaccine Efficacy Against the SARS-CoV-2 Omicron: A Systemic Review and Meta-Analysis

“…The included studies had a total sample of around one hundred million (N = 148260342) participants and of these, 74.2% were fully vaccinated ( n = 110076947) with at least two-doses of COVID-19 vaccines ( Table 1 ). Among the 13 studies, 1 study reported the effectiveness of the BNT162b2 (Pfizer–BioNTech), ChAdOx1 (AstraZeneca), mRNA-1273 (Moderna) and JNJ-78436735 vaccines ( 25 ); 1 study reported the effectiveness of the BNT162b2 and ChAdOx1 vaccines ( 24 ); 3 studies reported the effectiveness of the BNT162b2, ChAdOx1, and mRNA-1273 vaccines ( 12 , 21 , 22 ); 5 studies reported the effectiveness of the BNT162b2, and mRNA-1273 vaccines ( 14 , 16 , 17 , 20 , 23 ); 2 studies reported the effectiveness of the BNT162b2 vaccine ( 18 , 19 ), and 1 study reported the effectiveness of the mRNA-1273 vaccine ( 15 ).…”

“…In this analysis, the effectiveness of the vector vaccine (ChAdOx1) was reported by in five studies ( 12 , 21 , 22 , 24 , 25 ). The pooled effectiveness of the vector vaccine was unable to be evaluated as most of these studies did not report the effectiveness of each vaccine individually.…”

“…Figure 2A shows the effectiveness of the COVID-19 vaccines against the Omicron variant after at least two doses of the vaccination. All (12,(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)25) but one study (24) showed that vaccines played a significant role in reducing the risk of B.1.1.529 Omicron variant infection among the fully vaccinated population (two doses with or without boosters), compared to the unvaccinated population. The pooling of 13 studies showed that no significant difference between fully vaccinated population and unvaccinated population (OR = 0.77, 95% CI: 0.32-1.83).…”

“…The pooling of 13 studies showed that no significant difference between fully vaccinated population and unvaccinated population (OR = 0.77, 95% CI: 0.32-1.83). After excluding the one study (24) looking at the long term effectiveness of the vaccinated population, the pooling of the remaining 12 studies showed that full vaccination significantly lowered the risk (OR = 0.62, 95% CI: 0.56-0.69) of infection against the Omicron variant (Figure 2B). This data prompted us to examine the effectiveness of the booster vaccination.…”