The rapid global spread of the coronavirus disease (COVID-19) has compelled national governments to issue guidance on the use of face masks for members of the public. Growing evidence of transmission from asymptomatic and presymptomatic individuals makes the development of these guidelines increasingly pressing [1-3]. Recent research suggests that surgical masks could help prevent transmission of human coronaviruses by reducing emissions of coronavirus RNA in respiratory droplets and aerosols [4]. Although N95 respirators have the potential for even greater protection when compared to surgical masks [5,6], they also require fit testing that make them unsuitable for the public at large [7]. Currently, both surgical masks and N95 respirators are in short supply, with health care workers continuing to face shortages of personal protective equipment (PPE). In light of this, fabric masks have become a third masking option, although current evidence on their efficacy is limited. Prior studies suggest that fabric masks are significantly less effective than surgical masks, both for protecting health care workers and for reducing spread among the general public [8,9]. Even with lower efficacy, however, all masking options appear to hold value. Recent modeling suggests that widespread public adoption of even relatively ineffective masks would be able to help curtail community transmission of COVID-19, although more effective masks yield greater reductions in mortality [10].

Despite growing evidence on the value of masking and calls for public use of masks as part of a broader strategy that also includes social distancing and hand washing [10,11], recent commentary suggests that public guidance on masks may be inconsistent across nations [12], and the World Health Organization (WHO) maintains, as of May 2020, that masks are only needed for healthy individuals when they are taking care of someone with suspected COVID-19 [13]. Given the pandemic status of COVID-19, it is critical to establish a baseline understanding of current government guidelines on mask use for the general public. To date, no studies have conducted a systematic analysis of mask guidelines aimed at the public. Public-facing guidelines are critical to compliance since government provision of cues is an important driver of mask use [14]. Agency websites are a particularly critical way to disseminate these types of guidelines, as the public increasingly turns to the internet for health information.

To inform this discussion and help health agencies to “adopt rational recommendations on appropriate face mask use” [12], this paper presents a content analysis of health agency mask guidelines in March 2020 among countries and regions with large numbers of COVID-19 cases.

The 25 countries and regions with the highest number of confirmed COVID-19 cases were drawn from the Johns Hopkins Center for Systems Science and Engineering Coronavirus COVID-19 Global Cases tracker on March 9, 2020 [15]. These countries and regions are listed in Table 1. To replicate the experience of someone looking for guidance, we visited national health agency websites for each country and region seeking mask guidelines (medical, surgical, and unspecified mask types) aimed at the public. Given limited evidence for their efficacy relative to surgical and medical masks [8,9], we considered fabric masks separately. Specifically, we sought to find both recommendations for or against fabric masks for primary use and recommendations for fabric masks only when other more effective masks are unavailable. A content analysis approach was used [16], and a codebook was developed in Excel (Microsoft Corporation) to track guidance on when masks are recommended or not recommended. Initial coding suggested that several nations indicated that masks were not recommended because they may increase risks or create a false sense of safety. A code was also added to track these statements. Countries were assigned across the coding team by language proficiency, with Google Translate used as needed. When available, both the original and English language version of guidance were reviewed. All relevant webpages and documents were downloaded to create a static record. Websites were coded between March 13 and March 23, 2020, with all coding verified by a second coder the following week. Any coding discrepancies were discussed among authors and resolved. All websites were revisited a final time on March 30 to look for updated materials, and coding was updated as needed.

All 25 countries and regions had some form of publicly available information about masks on their health agency websites aimed at the public. Format and level of detail ranged greatly and included infographics (eg, Malaysia) and short responses in a frequently asked questions format (eg, Netherlands). Iraq had the vaguest guidance, which the coding team inferred from news stories and press releases about mask use rather than the existence of a formal page or document with public COVID-19 prevention guidance. A total of 4 (16%) countries and regions lacked recommendations for wearing surgical, medical, or unspecified masks when symptomatic, and 12 (48%) countries did not mention use by individuals providing care during home quarantine (Table 1). Although 9 (36%) countries and regions recommended surgical, medical, or unspecified masks in public or poorly ventilated places, 16 (64%) explicitly recommended against the general public wearing masks. A total of 7 (28%) also noted that surgical, medical, or unspecified masks were not recommended because they could increase health risks to the wearer or give a false sense of security.

With regard to fabric masks, no countries or regions recommended this mask type as preferable to surgical or medical masks in the Table 1 scenarios. Out of the 25 countries and regions, there was a country and a region (n=2, 8%) that explicitly recommended against them due to their protective capacity being either unknown (Italy) or inadequate (Hong Kong). South Korea and Mainland China recommended fabric masks as part of a broader guidance of different mask types being appropriate for different risk scenarios. South Korea, for example, noted: “In cases where there is not a high risk of infection or there is no health mask, it is helpful to use a cotton mask (including replacing the electrostatic filter) to avoid droplets directly from coughing or sneezing.” Germany, the United States, and Japan recommended some form of fabric mouth covering (including scarves and handkerchiefs) only when other options were unavailable. Austria was the only country or region at the time of analysis to recommend fabric masks interchangeably with other types of masks, noting that a “textile mouth-nose guard can also be used” as part of guidance on mask use in public spaces. The remaining 17 (68%) countries and regions did not explicitly address fabric masks in their guidance.

In some cases, countries or regions updated their guidance during the study period. For example, Iranian guidance initially recommended asymptomatic individuals not wear masks. By March 29, 2020, additional guidance was posted recommending masks at the park, gym, or when engaging in urban travel. On March 30, Austria removed guidelines referencing the WHO that stated that “disposable face masks are not an effective protection” and instead recommended “protective mask[s] in public spaces where there may be close contact with other people, e.g. in supermarkets.” By contrast, Sweden scaled back guidance during the study period, removing language that masks could help prevent spread from symptomatic individuals.