Excess Mortality in the United States During the First Three Months of the COVID-19 Pandemic

  1. R. Rivera
  2. J. E. Rosenbaum
  3. W. Quispe

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Deaths are frequently under-estimated during emergencies, times when accurate mortality estimates are crucial for emergency response. This study estimates excess all-cause, pneumonia, and influenza mortality during the COVID-19 pandemic using the September 11, 2020 release of weekly mortality data from the United States (U.S.) Mortality Surveillance System (MSS) from September 27, 2015 to May 9, 2020, using semiparametric and conventional time-series models in 13 states with high reported COVID-19 deaths and apparently complete mortality data: California, Colorado, Connecticut, Florida, Illinois, Indiana, Louisiana, Massachusetts, Michigan, New Jersey, New York, Pennsylvania, and Washington. We estimated greater excess mortality than official COVID-19 mortality in the U.S. (excess mortality 95% confidence interval (CI) (100013, 127501) vs. 78834 COVID-19 deaths) and 9 states: California (excess mortality 95% CI (3338, 6344) vs. 2849 COVID-19 deaths); Connecticut (excess mortality 95% CI (3095, 3952) vs. 2932 COVID-19 deaths); Illinois (95% CI (4646, 6111) vs. 3525 COVID-19 deaths); Louisiana (excess mortality 95% CI (2341, 3183) vs. 2267 COVID-19 deaths); Massachusetts (95% CI (5562, 7201) vs. 5050 COVID-19 deaths); New Jersey (95% CI (13170, 16058) vs. 10465 COVID-19 deaths); New York (95% CI (32538, 39960) vs. 26584 COVID-19 deaths); and Pennsylvania (95% CI (5125, 6560) vs. 3793 COVID-19 deaths). Conventional model results were consistent with semiparametric results but less precise. Significant excess pneumonia deaths were also found for all locations and we estimated hundreds of excess influenza deaths in New York.

We find that official COVID-19 mortality substantially understates actual mortality, excess deaths cannot be explained entirely by official COVID-19 death counts. Mortality reporting lags appeared to worsen during the pandemic, when timeliness in surveillance systems was most crucial for improving pandemic response.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2020.05.04.20090324: (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
    Among the 59 National Syndromic Surveillance Program (NSSP) jurisdictions, we were only able to obtain daily ED visits in New York City for asthma symptoms.
    Syndromic Surveillance Program
    suggested: None

    Results from OddPub: Thank you for sharing your code and data.

    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 10.1101/2020.05.04.20090324 on medRxiv
  3. Version published to 10.1017/s0950268820002617