Linguistic fairness in the U.S.: The case of multilingual public health information about COVID-19

  1. Damián E. Blasi
  2. Vishala Mishra
  3. Adolfo M. García
  4. Joseph P. Dexter

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Lack of high-quality multilingual resources can contribute to disparities in the availability of medical and public health information. The COVID-19 pandemic has required rapid dissemination of essential guidance to diverse audiences and therefore provides an ideal context in which to study linguistic fairness in the U.S. Here we report a cross-sectional study of official non-English information about COVID-19 from the Centers for Disease Control and Prevention, the Food and Drug Administration, and the health departments of all 50 U.S. states. We find that multilingual information is limited in many states, such that almost half of all individuals not proficient in English or Spanish lack access to state-specific COVID-19 guidance in their primary language. Although Spanish-language information is widely available, we show using automated readability formulas that most materials do not follow standard recommendations for clear communication in medicine and public health. In combination, our results provide a snapshot of linguistic unfairness across the U.S. and highlight an urgent need for the creation of plain language, multilingual resources about COVID-19.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2021.09.27.21264211: (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
    These corpora are large (the SUA contains approximately 3 billion tokens in total) and cover a range of document types, including Wikipedia and news articles, government documents, and transcripts of speeches.
    Wikipedia
    suggested: (Wikipedia, RRID:SCR_004897)

    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:
    Despite the well-known limitations of readability formulas, including focus on shallow linguistic features and often incomplete validation (32–35), we were able to confirm their reliability for relative assessment of text difficulty; in particular, all metrics considered yielded intuitively reasonable difficulty rankings across a diverse collection of Spanish prose. While our results identify points of urgent concern for the ongoing pandemic response, further work is needed to expand the scope of evaluation to include multimedia resources and traditional, offline media, as well as guidance issued by municipal and county health officials (39). Such an expanded evaluation should seek to integrate statistical and computational evidence with empirical assessments of comprehension and usability testing of multilingual resources (9, 40). Despite extensive efforts to address the COVID-19 “infodemic” (41), proliferation of misinformation, especially about COVID-19 vaccines, remains a persistent issue (42, 43). Our results thus point to a concerning combination of structural factors - incomplete coverage of multilingual resources, restricted availability of plain language materials, and, as documented previously, poor media ecology - that make it particularly challenging for U.S. residents with limited English proficiency to access trustworthy information. In light of the limitations of official information, many private groups have worked during the pandemic to improve health equity ...

    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.
    • No funding statement was detected.
    • 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.

  2. Version published to 10.1101/2021.09.27.21264211v1 on medRxiv