Many people throughout the world continue to debate whether the coronavirus disease 2019 (COVID-19) pandemic was deserving of the intensive and extensive efforts made by the government and private sectors alike to develop population-wide vaccines and vaccination programs. A new study posted on the medRxiv* preprint server examines hospitalization and mortality rates from COVID-19 before and after the rollout of the vaccine in England.
Study: Estimating the effectiveness of COVID-19 vaccination against COVID-19 hospitalisation and death: a cohort study based on the 2021 Census, England. Image Credit: Looker Studio / Shutterstock.com
*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.
About the study
Early in December 2020, the COVID-19 messenger ribonucleic acid (mRNA) vaccine from Pfizer/BioNTech was rolled out in the United Kingdom, followed by the Moderna mRNA and Oxford/AstraZeneca adenovirus vector vaccine ChAdOx1-S. Clinical trials on each of these vaccines showed high vaccine effectiveness (VE) in preventing symptomatic and severe COVID-19.
In the current study, researchers examine population-level data from the 2021 Census in England that, comprised over 580,000 individuals 16 years or older.
The data of individuals who had received one, two, or three doses of the vaccine were examined and stratified by the time since each dose. The primary outcome of this study was to assess the likelihood of hospitalization for vaccinated participants after each exposure and at different time points following each dose using unvaccinated participants as controls.
VE against hospitalization
One dose of the COVID-19 vaccine reduced the likelihood of hospitalization by 52%, whereas two doses increased VE to 56%.
VE against COVID-19 hospitalization was negative three or more months after the first dose, irrespective of age; however, it was highest in those between 30-64 years and 65-79 years of age after the second dose. Following the third booster dose, VE peaked in the 65-79 years age group at about 88%.
After three doses, the risk of hospitalization declined by almost 78%. The protection offered by a full primary course and third booster dose declined to about 68% at three or more months from the third dose. The greatest reduction in VE at this point was in those between 30-64 years of age.
VE against mortality
The risk of mortality declined by almost 60% after one vaccine dose. A notable exception is in those over the age of 80, who exhibited almost zero VE with one dose. Protection rose to 90% overall after the second dose; however, it remained lower in those over 80 years old as compared to the high VE in individuals aged 30-79 years.
The highest VE of 93% was observed after the booster dose; however, those between 16-29 years exhibited a VE of 71%. Moreover, VE declined at three or more months from the second and third doses but remained high except in the 16-29 years age group, where it waned substantially.
VE by variant
During the Omicron wave, the COVID-19 vaccine better protected individuals against hospitalization than previous reports; however, VE against mortality declined.
Moreover, VE against hospitalization was higher after the second dose than for either the first or third dose during the Omicron wave. However, VE declined at both three to six months and further reduced to about 40% over six months from the second dose.
After the third dose, protection against hospitalization declined from 80% before the onset of the Omicron wave to 55% thereafter. However, the reduction of VE by 23% that was observed three months after the third dose pre-Omicron did not occur during the Omicron wave.
Protection against COVID-19 mortality declined after the first, second, and third vaccine doses during the Omicron-dominant period. In fact, the first dose did not confer significant protection, whereas VE against mortality declined to about 50% over six months after the second dose.
VE by vaccine type
Overall, mRNA vaccines performed better than other COVID-19 vaccines; however, the population protected in each case varied. The mRNA vaccines showed waning VE against hospitalization at three months or more from the booster dose, whereas both vaccine types exhibited waning VE against mortality three months or more following the second dose.
Confounding factors
Confounding factors likely affected the calculation of VE against mortality due to non-COVID-19 causes.
Some possible causes of error could be the administration of vector-based vaccines to the most frail and at-risk patients and the vaccination of healthy individuals with more effective vaccines. The follow-up in this study began in late March 2021, whereas the vaccination campaign began on December 8, 2022.
Thus, the oldest individuals in this study were likely frailer or lacked access to vaccination sites at the beginning of the campaign. This affected their chances of developing severe or fatal COVID-19 after the first dose.
Frailty biases likely explain the lower VE after the first and second doses, as these were probably older and sicker people whose vaccination was delayed due to health concerns.
Similarly, younger and healthy people are likelier to have received the booster doses later. As a result, the three to six and over six-month follow-up after the booster dose will be biased towards high-risk individuals. These findings indicate that health status is a significant confounding factor after adjusting for other factors.
Given the inconsistency of the decrease across the breakdowns, we cannot say there is strong evidence of waning protection against COVID-19 hospitalisation. For VE against COVID-19 mortality, we do see consistent decreases across the various breakdowns. This could indicate waning of protection.”
What are the implications?
The vaccine effectiveness estimates show increased protection with number of doses and a high level of protection against both COVID-19 hospitalisation and mortality for the third/booster dose.”
For the first time, the current study adjusted for socioeconomic and health status variables using data from the 2021 Census, thereby improving the accuracy of real-world VE estimation. Nevertheless, residual confounding by health status is observable, which should be factored into any VE estimation tool in future studies.
*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.
Research links COVID-19 vaccines to temporary facial palsy in over 5,000 patients