The Causal Effect of Air Pollution on COVID-19 Transmission: Evidence from China

  1. Guojun He
  2. Yuhang Pan
  3. Takanao Tanaka

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

There is increasing concern that ambient air pollution could exacerbate COVID-19 transmission. However, estimating the relationship is challenging because it requires one to account for epidemiological characteristics, to isolate the impact of air pollution from potential confounders, and to capture the dynamic impact. We propose a new econometric framework to address these challenges: we rely on the epidemiological Susceptible-Infectious-Recovered-Deceased (SIRD) model to construct the outcome of interest, the Instrument Variable (IV) model to estimate the causal relationship, and the Flexible-Distributed-Lag (FDL) model to understand the dynamics. Using data covering all prefectural Chinese cities, we find that a 10-point (14.3%) increase in the Air Quality Index would lead to a 2.80 percentage point increase in the daily COVID-19 growth rate with 2 to 13 days of delay (0.14 ∼ 0.22 increase in the reproduction number: R 0 ). These results imply that improving air quality can be a powerful tool to contain the spread of COVID-19.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2020.10.19.20215236: (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

    No key resources detected.

    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: We detected the following sentences addressing limitations in the study:
    Our study has two caveats. First, we use confirmed active infections to create our outcome variables. As is common in any study of infectious disease39, the number of confirmed cases could be much lower than the actual cases, and confirmed cases might not reflect the real epidemic outbreak. This concern is partially alleviated by our use of the growth rate as the outcome variable because our results will hold as long as under-reporting is constant (for example, if 50% of infected cases are always confirmed) within a city. In addition, our regressions include date fixed effects that can control for nationwide events specific to each date, such as national testing policies or revision of the disease classification. While controlling the variation in testing capacity can partially mitigate the concern about under-reporting, such data are not available at the city-by-day level. Second, we do not have sufficient statistical power to investigate whether air pollution affects COVID-19 deaths. This is because there have been only a few COVID-19 deaths in most Chinese cities. Outside Wuhan, more than 90% of Chinese cities have recorded only 0 or 1 death. While more than 3,000 people died from COVID-19 in Wuhan city, the data were not accurate at the early stage of the outbreak in the city39. Therefore, we refrain from discussing the relationship between air pollution and the COVID-19 death rate. Nevertheless, we investigate this issue in Supplementary Figure 9 and Supplementary Note 4...

    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 10.1101/2020.10.19.20215236v1 on medRxiv