Evolution of case fatality rates in the second wave of coronavirus in England: effects of false positives, a Variant of Concern and vaccination

  1. James A Ackland
  2. Graeme J Ackland
  3. David J Wallace

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

1

Objective

To track the statistical case fatality rate (CFR) in the second wave of the UK coronavirus outbreak, and to understand its variations over time.

Design

Publicly available UK government data and clinical evidence on the time between first positive PCR test and death are used to determine the relationships between reported cases and deaths, according to age groups and across regions in England.

Main Outcome Measures

Estimates of case fatality rates and their variations over time.

Results

Throughout October and November 2020, deaths in England can be broadly understood in terms of CFRs which are approximately constant over time. The same CFRs prove a poor predictor of deaths when applied back to September, when prevalence of the virus was comparatively low, suggesting that the potential effect of false positive tests needs to be taken into account. Similarly, increasing CFRs are needed to match cases to deaths when projecting the model forwards into December. The growth of the S gene dropout VOC in December occurs too late to explain this increase in CFR alone, but at 33% increased mortality, it can explain the peak in deaths in January. Seasonal effects could be in part responsible for the early December increase in CFR, and if so, the estimate of increased mortality would be reduced. There is also evidence that the prevalence of B.1.1.7 may have been slower amongst older age groups, and if this is a factor, then 33% could be an underestimate of mortality. From the second half of January, the CFRs for older age groups show a marked decline. Since the fraction of the VOC has not decreased, this decline is likely to be the result of the rollout of vaccination. However, due to the rapidly decreasing nature of the raw cases, any imprecisions in the time-to-death distribution are magnified in this time period, rendering estimates of vaccination’s effect less precise.

Conclusions

The relationship between cases and deaths, even when controlling for age, is not static through the second wave of coronavirus in England. An apparently anomalous low case-fatality ratio in September can be accounted for by a 0.4% false-positive fraction. The rapid growth in CFR in December can be understood in part in terms of a more deadly new variant B.1.1.7, while a decline in January correlates with vaccine roll-out, suggesting that vaccine reduce the severity of infection, as well as the risk.

Summary Box

What is already known on this topic

The case fatality rate (CFR) is a useful measure which enables one to estimate future deaths based on current infections. In England, there was a surge in Covid-19 CFR around the beginning of December.

What the study adds

Using it, we monitor the case-fatality rate across time, region and age group from publicly available case data. This quantity is related to the lethality of the virus. It shows a sharp increase in December 2020, which parallels the spread of the B.1.1.7 variant. The January peak in actual deaths matches that predicted by cases if B.1.1.7 is about 33% more deadly; this estimate would be lower if there is a seasonal effect on deaths, and higher if at the peak the variant was less pervasive amongst older age groups. A steady drop in CFR from January suggests that vaccination not only reduces transmission but also the risk of serious illness among those infected. It is notable that these conclusions are reached with publicly available data independent of clinical studies.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2021.04.14.21255385: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    NIH rigor criteria are not applicable to paper type.

    Table 2: Resources

    Software and Algorithms
    SentencesResources
    Analysing cases and deaths data for England for each of the age groups enables us to determine CFRs for each age group, without depending on the values inferred from the CDC study [3] in [2], which may no longer apply to the new variants and circumstances of the pandemic in England. 3.1 Distributions of time to death: To identify an appropriate distribution to improve the ‘shift’ assumption, we have reviewed six studies reporting time to death distributions [9-14].
    England.
    suggested: None
    Calculations were performed independently in R and in Excel, providing an element of code verification.
    Excel
    suggested: None

    Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).

    Results from LimitationRecognizer: An explicit section about the limitations of the techniques employed in this study was not found. We encourage authors to address study limitations.

    Results from TrialIdentifier: No clinical trial numbers were referenced.

    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

    Results from JetFighter: We did not find any issues relating to colormaps.

    Results from rtransparent:
    • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
    • Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
    • No protocol registration statement was detected.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  2. Version published to 10.1101/2021.04.14.21255385v1 on medRxiv