Occupational differences in SARS-CoV-2 infection: analysis of the UK ONS COVID-19 infection survey

  1. Sarah Rhodes
  2. Jack Wilkinson
  3. Neil Pearce
  4. William Mueller
  5. Mark Cherrie
  6. Katie Stocking
  7. Matthew Gittins
  8. Srinivasa Vittal Katikireddi
  9. Martie Van Tongeren

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Concern remains about how occupational SARS-CoV-2 risk has evolved during the COVID-19 pandemic. We aimed to ascertain occupations with the greatest risk of SARS-CoV-2 infection and explore how relative differences varied over the pandemic.

Methods

Analysis of cohort data from the UK Office of National Statistics COVID-19 Infection Survey from April 2020 to November 2021. This survey is designed to be representative of the UK population and uses regular PCR testing. Cox and multilevel logistic regression were used to compare SARS-CoV-2 infection between occupational/sector groups, overall and by four time periods with interactions, adjusted for age, sex, ethnicity, deprivation, region, household size, urban/rural neighbourhood and current health conditions.

Results

Based on 3 910 311 observations (visits) from 312 304 working age adults, elevated risks of infection can be seen overall for social care (HR 1.14; 95% CI 1.04 to 1.24), education (HR 1.31; 95% CI 1.23 to 1.39), bus and coach drivers (1.43; 95% CI 1.03 to 1.97) and police and protective services (HR 1.45; 95% CI 1.29 to 1.62) when compared with non-essential workers. By time period, relative differences were more pronounced early in the pandemic. For healthcare elevated odds in the early waves switched to a reduction in the later stages. Education saw raises after the initial lockdown and this has persisted. Adjustment for covariates made very little difference to effect estimates.

Conclusions

Elevated risks among healthcare workers have diminished over time but education workers have had persistently higher risks. Long-term mitigation measures in certain workplaces may be warranted.

Article activity feed

  1. Version published to 10.1136/jech-2022-219101
  2. Version published to 10.1101/2022.04.28.22273177v2 on medRxiv
  3. ScreenIT

    SciScore for 10.1101/2022.04.28.22273177: (What is this?)

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

    Table 1: Rigor

    Ethicsnot detected.
    Sex as a biological variablenot detected.
    RandomizationDataset: The CIS is a randomly sampled repeated cross-sectional household survey with serial sampling designed to be representative of the UK population.
    Blindingnot detected.
    Power Analysisnot detected.

    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:
    Limitations: The ONS infection survey was a prevalence study which in the most part conducted monthly tests for participants; it is likely that positive results were missed in between visits. While this is relevant to prevalence estimates, it is less likely to affect relative effects. The CIS started in April 2020, several months into the pandemic, so we are likely to have missed a period where occupational risks would have been most evident for some occupational groups. It is possible that certain occupations will be underrepresented in the survey due to their availability for the study visits because of long working hours or shift work. There is risk of selection bias; e.g. healthcare workers who were front-line may have been less likely to be recruited or less likely to provide data than those who were non-frontline due to shift work or lack of time. Occupational information, particularly 4-digit SOC, was missing for a proportion of participants. We used sensitivity analyses and also accompanied analyses by occupation with analysis by sector (where information was more complete) in an attempt to check that our findings are robust. There is likely to be variation within occupational groups and sectors that will be masked when assessing group averages. The sample is not large enough to make meaningful analysis of more granular groupings, particularly when assessing separate time periods.

    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.

    Results from scite Reference Check: We found no unreliable references.

    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.

  4. Version published to 10.1101/2022.04.28.22273177v1 on medRxiv