COVID-19 Science Update released: March 26, 2021 Edition 82

COVID-19 Science Update

From the Office of the Chief Medical Officer, CDC COVID-19 Response, and the CDC Library, Atlanta GA. Intended for use by public health professionals responding to the COVID-19 pandemic.

*** Available on-line at ***


Sign up for COVID-19 Science Updates

Section headings in the COVID-19 Science Update align with the CDC Science Agenda for COVID-19.

Section headings in the COVID-19 Science Update have been changed to align

with the CDC Science Agenda for COVID-19.

PDF version for this update pdf icon[PDF – 865 KB]

COVID-19 Science Updates

Here you can find all previous COVID-19 Science Updates.

Prevention, Mitigation and Intervention Strategies


Vaccination and non-pharmaceutical interventions for COVID-19: a mathematical modelling studyexternal icon. Moore et al. The Lancet Infectious Diseases (March 18, 2021).

Key findings:

  • R is projected to remain >1 under all scenarios of vaccine efficacy scenarios (Figure 1):
    • With the rollback of NPIs and most optimistic (85%) efficacy against infection, R = 1.58 (95% credibility interval [CI] 1.36-1.83).
  • Projected COVID-19 deaths under default uptake scenarios and removal of NPIs once all adults have received 2 vaccine doses:
    • 21,400 deaths (95% CI 1,400-55,100) for a vaccine that prevents 85% of infections.
    • 96,700 deaths (95% CI 51,800-173,200) for a vaccine that prevents 60% of infections.

Methods: A “susceptible-exposed-infected-recovered” (SEIR) model was fitted to UK epidemiological data to evaluate the effect of the UK vaccine rollout combined with lifting non-pharmaceutical interventions (NPI) to predict the reproduction number (R), COVID-19 daily deaths and hospital admissions between January 2021 and January 2024. Default assumptions for vaccine efficacy for symptomatic disease was 88% and vaccine uptake by age ranged from 75%–95%. The model also considered vaccine efficacy against infection which was varied from 0 to 85% and the spread of B.1.1.7. Limitations: Not considered: effects of waning immunity; variants of concern (VOCs) that escape immune response; varying vaccine efficacy by age.

Implications: Vaccination alone is unlikely to contain SARS-CoV-2. Gradual easing of non-pharmaceutical interventions in addition to high population uptake of a high-efficacy vaccine are necessary to prevent subsequent waves of infection. Monitoring of VOCs such as those that escape immune response (e.g., E484K mutation) will be necessary.


Note: Adapted from Moore et al. Estimated reproduction number (R) after NPI release for a given number of administered doses for 4 vaccine scenarios with varying infection prevention efficacy: 0%, 35%, 60%, and 85%. Model excludes any additional increase in immunity from natural infection (after 1/29/2021) and any effect on contact patterns resulting from NPIs. Licensed under CC-BY-NC-ND.

Impact of the COVID-19 vaccine on asymptomatic infection among patients undergoing pre-procedural COVID-19 molecular screeningexternal icon. Tande et al. Clinical Infectious Diseases (March 10, 2021).

Key findings:

  • Vaccinated persons had a 65% reduction in risk for asymptomatic SARS-CoV-2 infection during a pre-procedure screening compared to unvaccinated patients (adjusted relative risk [aRR] = 0.35, 95% CI 0.26-0.47).
  • Vaccinated persons had an 80% reduction in risk for asymptomatic SARS-CoV-2 infection after either (1) testing >10 days post first dose (aRR = 0.21, 95% CI 0.12-0.37) or (2) testing after 2 doses (aRR = 0.20, 95% CI 0.09-0.44) of mRNA vaccine, compared to unvaccinated persons.

Methods: Retrospective cohort study of consecutive, adult patients (N = 39,156) asymptomatic for COVID-19 within a large US healthcare system who underwent pre-procedural SARS-CoV-2 screening between December 17, 2020, and February 8, 2021. Primary exposure was vaccination with at least one dose of Pfizer-BioNTech or Moderna vaccine prior to screening (median time from first dose to screening = 16 days). Adjusted relative risk (aRR) of positive SARS-CoV-2 molecular test was calculated adjusting for age, sex, race/ethnicity, and patient residence in Health Referral Region (local vs. non-local). Limitations: Unmeasured confounding among vaccinated individuals; generalizability may be limited as study participants were largely non-Hispanic white and <65 years; no longitudinal follow-up on symptom status post-procedure.

Implications: This study provides real-world evidence of a decrease in asymptomatic infection after vaccination with a mRNA vaccine and should be considered in developing post-vaccination policy recommendations.



Efficacy of the ChAdOx1 nCoV-19 Covid-19 vaccine against the B.1.351 variant.external icon Madhi et al. NEJM (March 16, 2021).

Key findings:

  • Primary end-point analysis: Overall vaccine efficacy against mild-to-moderate COVID-19 was 21.9% (95% CI -49.9%-59.8%).
  • Secondary end-point analysis among participants who developed mild to moderate COVID-19, 39 (92.9%) were infected with B.1.351:
    • Efficacy of a 2-dose ChAdOx1 nCoV-19 vaccine regimen against the B.1.351 variant was not evident (efficacy = 10.4%, 95% CI: -76.8%-54.8%).
  • The vaccine induced strong neutralizing antibody levels against original D614G virus on pseudovirus neutralization assay 14 days after second vaccine dose.

Methods: Double-blind, multicenter RCT in South Africa (between June 24 and November 9, 2020) was conducted to assess ChAdOx1 nCoV-19 (AstraZeneca-Oxford), primary efficacy analysis for vaccine efficacy (vaccine n = 750, placebo n = 717). Efficacy against B.1.351 analyzed in secondary endpoint analysis among 42 participants who developed COVID-19. Serum samples collected from 25 persons after having received the second vaccine dose were tested for neutralizing activity. Limitations: Lack of severe COVID-19 cases prevents conclusions around severe illness; limited generalizability.

Implications: The lack of efficacy of ChAdOx1 nCoV-19 against mild to moderate COVID-19 associated with the B.1.351 variant suggests that second generation vaccinations or boosters may be needed to address emerging variants of concern.

Detection, Burden, and Impact
Increased Mortality Associated with B.1.1.7 Variant


The SARS-CoV-2 variant of concern B.1.1.7 that was first identified in the UK in the fall of 2020 has become the dominant strain in the UK and parts of Europe and is increasing in prevalence across the US. The B.1.1.7 variant has been associated with increased transmission compared to the wild type lineage. The following two studies present findings regarding increased mortality of B.1.1.7.

A. Increased mortality in community-tested cases of SARS-CoV-2 lineage B.1.1.7external icon. Davies et al. Nature (March 15, 2021).

Key findings:

  • Prevalence of spike gene target failure (SGTF), associated with B.1.1.7 infection, increased steeply over time, from 5.8% in November 2020 to 94.3% in February 2021 (Figure panel a).
  • 1.1.7 is associated with increased mortality compared to other SARS-CoV-2 variants (hazard ratio = 1.61, 95% CI 1.42-1.82), after correcting for missing data/misclassification of SGTF weighting). (Figure panels b, c)

Methods: Analysis of a dataset containing 2,245,263 positive SARS-CoV-2 PCR tests and 17,642 COVID-19 deaths from community-based testing in England between September 1, 2020 and February 14, 2021. Identification of the B.1.1.7 variant in infected individuals was confirmed by spike gene target failure status (SGTF). Absolute mortality risks and hazard of death were calculated after confirmation of SGTF status in numerous demographic subsets. Limitations: Could not completely assess SGTF status within study population; non-random sample.



Note: Adapted from Davies et al. a) The number of samples with SGTF and without SGTF by day from November 1, 2020–February 14, 2021; b) Number of deaths within 28 days of positive test by specimen date with SGTF and without SGTF; and c) Kaplan-Meier plot showing survival (95% confidence intervals) among individuals who were community-tested with SGTF and without SGTF. Permission request in process.

B. Case fatality risk of the SARS-CoV-2 variant of concern B.1.1.7 in England, 16 November to 5 February.external icon Grint et al. Eurosurveillance (March 18, 2021).

Key findings:

  • Risk of death was greater (adjusted hazard ratio 1.67, 95% CI 1.34-2.09) among individuals with confirmed B.1.1.7 variant of concern compared with individuals with non-1.1.7 SARS-CoV-2.
  • 28 days after a SARS-CoV-2-positive test, individuals with B.1.1.7 had a higher absolute risk of death compared with non-B.1.1.7 SARS-CoV-2 in all groups stratified by sex, age, and presence of comorbidities (Figure):
    • Risk of death was higher for males and increased with age and comorbidities.
    • Those ≥85 years with ≥2 comorbidities had the highest risk of death.

Methods: Cohort study among (n = 184,786) unvaccinated persons testing positive for SARS-CoV-2 between November 16, 2020, and January 11, 2021, in England. Spike gene target failure was used as a proxy for B.1.1.7 identification. Limitations: Selection bias might overestimate absolute mortality risks (if people with mild or asymptomatic infections were less likely to be tested); lead-time bias might underestimate relative mortality associated with B.1.1.7 given this variant was more prevalent later in the study period.


Note: Adapted from Grint et al. Excess risk of death by 28 days after positive test comparing SARS-CoV-2 VOC infection with non-VOC infection among males (A); and among females (B). Licensed under CC BY 4.0.

Implications for both studies (Davies et al. and Grint et al.): Both studies were consistent with Challen et alexternal icon., regarding the magnitude (~60%–70%) of the increased risk of mortality associated with the B.1.1.7 variant compared with other SARS-CoV-2 lineages such as the wild type. Other analyses indicating increased transmissibility of B.1.1.7 as well as the increasing spread of B.1.1.7 within the US highlight the need for rapid implementation of vaccination and continued implementation of non-pharmaceutical mitigation measures.


A potential SARS-CoV-2 variant of interest (VOI) harboring mutation E484K in the spike protein was identified within lineage B.1.1.33 circulating in Brazil.external icon Resende et al. bioRxiv (March 13, 2021).

Key findings:

  • A new SARS-CoV-2 variant of interest (VOI) N.9, within lineage B.1.1.33, with 4 non-synonymous lineage-defining mutations (NSP3:A1711V, NSP6:F36L, S:E484K, and NS7b:E33A), was detected in Brazil between November 2020 and February 2021.
  • N.9 probably emerged in the Summer of 2020 (estimated to be August 15, 2020; 95% High Posterior Density June 16–September 22, 2020), and has been identified in 10 different Brazilian states (Figure).

Methods: The nextclade toolexternal icon was used to identify B.1.1.33 lineage mutations among SARS-CoV-2 samples from Brazil sequenced by the Fiocruz COVID-19 Genomic Surveillance Network between March 12, 2020, and January 27, 2021; or that were available in Global Initiative on Sharing All Influenza Dataexternal icon as of March 1, 2021. Used Bayesian reconstruction to estimate timing of VOI N.9 emergence. Limitations: Effects of VOI N.9 on transmission, severity, diagnosis, and treatment of SARS-CoV-2 are unknown.

Implications: The E484K mutation has been associated with resistance to monoclonal antibodies and reduced neutralization potency of sera from convalescent and vaccinated individuals. The rapid dispersion of VOI N.9 across Brazil and the presence of the E484K mutation warrant monitoring for evidence of impact on effectiveness of treatments and vaccines, and surveillance for further spread, including in the US.


Note: Adapted from Resende et al. Geographic distribution of the VOI N.9 identified in Brazil (states are abbreviated per the ISO standard). Licensed under CC-BY-ND 4.0.

Natural History of SARS-CoV-2 Infection


Assessment of protection against reinfection with SARS-CoV-2 among 4 million PCR-tested individuals in Denmark in 2020: a population-level observational study.external icon Hansen et al. The Lancet (March 17, 2021).

Key findings:

  • Overall protection against repeat infection for those previously infected was 78.8% (95% CI 74.9-82.1%).
    • Protection was 47.1% (95% CI 24.7-62.8) among those aged ≥65 years, 81.3% (95% CI 72.6-87.3) among those aged 50–64 years, 80.1% (95% CI 71.8-85.9) among those aged 35–49 years, and 82.7% (95% CI 77.1-86.9) among those aged 0-34 years.
    • There were no differences in in estimated protection against repeat infection by sex or evidence of waning protection over time (3–6 ≥7 months of follow-up).

Methods: Analysis of Danish population-level cohort (n = 2,432,509) tested for SARS-CoV-2 by RT-PCR between February 26 and December 31, 2020. Calculated rate ratios (RRs) for persons testing positive among those with a positive or negative test ≥3 months earlier, adjusted for potential confounders. Estimated protection against repeat infection (1-RR). Findings pertain to the alternative cohort analysis that includes data from both surges.  Limitations: Potential selection bias in obtaining testing based on presumed risk; misclassification based on imperfect sensitivity and specificity of RT-PCR tests or on persistent infections counted as reinfections; unknown applicability of findings to reinfections with new SARS-CoV-2 variants.

Implications: Findings support vaccination of previously infected individuals because natural protection is robust but not absolute. Persons ≥65 years, even if previously infected, should receive enhanced protective measures including vaccination.



Household SARS-CoV-2 transmission and children: a network prospective study.external icon Soriano-Arandes et al. Clinical Infectious Diseases (March 12, 2021).

Key findings:

  • SARS-CoV-2 transmission occurred in 62% of household contacts.
    • 80 (7.7%) children were household index cases (Figure).
    • 756 (72%) pediatric COVID-19 cases were secondary to an adult case.
  • Secondary attack rate (SAR) was significantly lower in households with pediatric vs. adult index cases (59.0% vs. 67.6%, p = 0.006).
    • Among pediatric index cases, SAR was lower during the school period vs. the summer period (53.0% vs. 64.4%, p = 0.02).

Methods: Spanish prospective observational study of SARS-CoV-2 household transmission in children (<16 years) with COVID-19 (n = 1,040) and household contacts (n = 3,392); data were collected between July 1 and October 31, 2020. Index case was defined as first member of the household to develop symptoms or test positive for SARS-CoV-2. Two study periods were established, summertime (July 1–September 15) and school time (September 16–October 31). Limitations: Pediatric cases did not include children ≥16 years; index case definition may lead to underestimation of contribution of asymptomatic individuals; findings may be dependent of variant distribution at time of study.

Implications: Children, whether symptomatic or not, do not greatly contribute to household clusters of infection and are unlikely to be major drivers of the pandemic even if attending school.


Note: Adapted from Soriano-Arandes et al. Percentage of positive SARS-CoV-2 RT-PCRs (pediatric vs. adult) in household by the age of the index cases. Permission request in process.


An observational cohort study on the incidence of SARS-CoV-2 infection and B.1.1.7 variant infection in healthcare workers by antibody and vaccination status.external icon Lumley et al. medRxiv (March 12, 2021).

Key findings:

  • Either 2 vaccine doses or natural infection provided ≥85% protection against symptomatic and asymptomatic SARS-CoV-2 infection in healthcare workers (Figure):
    • 2 vaccine doses: 90% lower incidence (0.10 adjusted incident rate ratio [aIRR], 95% CI 0.02-0.38); natural infection: 85% lower incidence (aIRR 0.15, 95% CI 0.08-0.26).
  • Single-dose vaccination reduced the incidence of symptomatic infection by 67% (aIRR 0.33, 95% CI 0.21-0.52) and any PCR-positive result by 64% (aIRR 0.36, 95% CI 0.26-0.50).
  • No evidence of differences in immunity induced by natural infection and vaccination for infections with S-gene target failure and the circulating variant, B.1.1.7.

Methods: Longitudinal cohort study of 1 versus 2 dose vaccine effectiveness among healthcare workers in England (n = 13,109) between April 23, 2020, and February 28, 2021. Vaccines include BNT162b2 (Pfizer-BioNTech; n = 8,285) and ChAdOx1 nCOV-19 (Oxford AstraZeneca; n = 2,738). The Thermo Fisher TaqPath PCR diagnostic assay was used to determine gene target failure, indicative of B.1.1.7. Incident rate calculated using Poisson regression model, adjusting for age, sex, occupational role, and temporal changes in incidence. Limitations: Findings might not be generalizable to the population; priority was given to staff who were at greater risk of SARS-CoV-2 infection; limited power to detect differences between vaccines.

Implications: Natural infection resulting in detectable antibodies and two vaccine doses provide protection against SARS-CoV-2 (re)infection, including against the B.1.1.7 variant. These findings could help inform post vaccination policies.

Note: Adapted from Lumley et al. Protection from infection by antibody and vaccination status, compared with seronegative individuals. Outcomes were either any PCR positive result or symptomatic PCR positive infection. The number of healthcare workers in each follow up group is shown. 95% confidence intervals are plotted, except for previously seronegative HCWs vaccinated twice who had no symptomatic PCR confirmed infections. Licensed under CC-BY 4.0.

In Brief

Detection, Burden, and Impact

Natural History of SARS-CoV-2 Infection


Note: Adapted from Long et al. Schematic showing structural changes present in the spike protein domains of major SARS-CoV-2 variants (S1-NTD, S1-RBD, S1, and S2). Reprinted from The American Journal of Pathology, online ahead of print. Long et al, Sequence analysis of 20,453 SARS-CoV-2 genomes from the Houston metropolitan area identifies the emergence and widespread distribution of multiple isolates of all major variants of concern. Copyright ©2021, with permission from Elsevier Inc. on behalf of the American Society for Investigative Pathology.


Note: Adapted from Collier et al. Neutralization potency of mRNA vaccine sera (pre-SARS-CoV-2 B.1.1.7) against pseudovirus bearing spike mutation in the B.1.1.7 lineage and B.1.1.7lineage with E484K mutation in the receptor binding domain. Vaccine dose 1 (n = 37) and vaccine dose 2 (n = 21) compared to wild-type. Wilcoxon matched pairs signed rank test p-values: **<.01, ***<.0001, ****<0.0001). Permission request in process.

Protection in Healthcare and Non-Healthcare Work Settings


Reproduced with permission from JAMA Surgery. 2021. Published online March 11, 2021. doi:10.1001/jamasurg.2021.0836. Copyright©2021 American Medical Association. All rights reserved.

Prevention, Mitigation, and Intervention Strategies

Disclaimer: The purpose of the CDC COVID-19 Science Update is to share public health articles with public health agencies and departments for informational and educational purposes. Materials listed in this Science Update are selected to provide awareness of relevant public health literature. A material’s inclusion and the material itself provided here in full or in part, does not necessarily represent the views of the U.S. Department of Health and Human Services or the CDC, nor does it necessarily imply endorsement of methods or findings. While much of the COVID-19 literature is open access or otherwise freely available, it is the responsibility of the third-party user to determine whether any intellectual property rights govern the use of materials in this Science Update prior to use or distribution. Findings are based on research available at the time of this publication and may be subject to change.


CDC logo /

Page last reviewed: March 26, 2021, 12:00 AM
Content source: Office of the Chief Science Officer - COVID-19