To Mask or Not to Mask? Let’s Look at the Science

To Mask or Not to Mask? Let’s Look at the Science

It’s a beautiful sunny day here in New York City. Sick of being cooped up in quarantine for so long, the sweet siren song of the sun now serenades us. Central Park beckons.

We all want nothing more than to go outside and feel the sun on our faces. But the pandemic looms on the horizon, hovering at the edge of our thoughts and impacting our daily lives in dozens of subtle or not-so-subtle ways. The question remains: will we finish out the summer with ridiculous looking mask-tans?

Do masks even help?

The topic of mask-wearing has been surrounded by controversy since the start of the current pandemic.

Way back in February (years ago in COVID-time) we saw the following tweet from the U.S. Surgeon General:

His well-intentioned comment aimed to preserve the limited supply of PPE (Personal Protective Equipment) for the healthcare workers (HCW) who needed it most. It also reflected the U.S. scientific consensus at the time on the effectiveness of mask-wearing by the general public. Unfortunately, that scientific consensus was based on a paucity of information.

Modern civilization has never faced a global pandemic of this scale. We have never had cause for entire populations to wear face-masks, so up until now, we haven’t had any large-scale data on their efficacy. Instead, we’ve been relying on mechanistic laboratory studies and small observational studies of healthcare workers dealing with influenza patients or previous outbreaks. These studies are helpful for understanding the following:

  • The ideal filtration characteristics of different types of masks and fabrics
  • The effectiveness of medical-grade masks in high-risk settings when worn by healthcare personal trained in how to use them.

The results of these studies are enough to provide hope for mask usage helping in a wider context, but they leave many questions unanswered. The one on everyone’s mind is as follows:

How effective are non-medical grade masks made out of cloth or other materials, worn by the general population untrained in proper mask usage, in contexts of unknown exposure levels?

The WHO conducted a systematic review of 172 observational studies across multiple settings and continents, including over 25,000 patients with COVID-19, SARS, or MERS. Of the 44 studies included in the final analysis, 7 were of COVID-19, 6 of which were conducted in China. They came to some important conclusions:

  1. Face-mask wearing reduced the risk of infection by 82%.
  2. This effect was stronger in healthcare settings, where N95 masks offered a 96% protection, vs 66% from surgical masks (this difference was not as reliable).
  3. Social distancing was shown to be highly effective. The risk of transmission went from ~12.8% for those who kept less than 1m (~3 feet) from others to 2.6% for those who kept more than that. There was an additional halving of risk for each extra meter of distance.

This study had several limitations. First, it only looked at N95 respirators or similar and surgical or similar masks (eg, 12–16-layer cotton or gauze masks). Second, the non-healthcare settings studied had relatively small sample sizes. Third, it only looked at studies published as of May 3rd, which themselves must have looked at data from previous months (hence most of the COVID studies coming from China).

Thankfully, we have finally reached a point where large-scale, COVID-specific data has become available.

In this article, we will examine real-world data and a large selection of studies either published or posted to pre-print servers over the last 5 weeks. We will attempt to answer a single question: does universal mask-wearing by the general populace help lower the spread of COVID-19?

Note: pre-print studies have not been peer-reviewed, and therefore should be evaluated with a critical eye.

The Current Stance of the WHO

On June 5th, the WHO finally updated its guidance document, “Advice on the use of masks in the context of COVID-19.” They recommend that all healthcare workers or people displaying symptoms should wear masks, but state that:

“At present, there is no direct evidence (from studies on COVID- 19 and in healthy people in the community) on the effectiveness of universal masking of healthy people in the community to prevent infection with respiratory viruses, including COVID-19.”

Seems pretty clearcut, no? Until they start to hedge:

“However, taking into account the available studies evaluating pre- and asymptomatic transmission, a growing compendium of observational evidence on the use of masks by the general public in several countries, individual values and preferences, as well as the difficulty of physical distancing in many contexts, WHO has updated its guidance to advise that to prevent COVID-19 transmission effectively in areas of community transmission, governments should encourage the general public to wear masks in specific situations and settings as part of a comprehensive approach to suppress SARS-CoV-2 transmission (Table 2).” (Bolding added by me)

These specific situations and settings essentially boil down to:

  1. If you are a medical worker or are displaying symptoms, wear a medical mask.
  2. If you have no symptoms but will be in a setting where distancing cannot be maintained, wear a non-medical mask as source control just in case you are an asymptomatic carrier.
  3. If you are over 60 years old or have comorbidities and will be in a setting where distancing cannot be maintained, wear a medical mask for protection.

Of the listed “Potential benefits” of population mask-wearing, the only medical item is: “reduced potential exposure risk from infected persons before they develop symptoms.”

This is based on the assumption that the disease primarily spreads through large droplets and direct contact, and even poor quality masks can block the big stuff.

In my next article, we will go over the technical aspects of disease transmission, the different types of masks, and what should be worn when for maximal effect. For now, let’s keep things simple. Most countries with mask-wearing requirements or guidelines allow any type of mask, so we should be able to look at the data and see if even this basic level of maskpertise helps to slow the progress of the disease.

Let’s look at some numbers.

Real-World Data: Does Mask Wearing Help?

In short: Yes.

The difficulty with most real-world studies is the impossibility of isolating cause and effect. If disease transmission in a country is seen to drop off after mask-wearing policies were instituted, but this happened at the same time as social distancing guidelines were released, it’s difficult to tell which intervention produced the observed effect. To circumvent this problem we typically use randomized controlled trials (RCTs) as the gold standard for determining cause and effect. Unfortunately, in the case of mask-wearing for a pandemic, RCTs are almost impossible to carry out, facing both ethical and practical difficulties.

Fortunately, there are other ways of isolating variables, especially in the case of large sample sizes and staggered policy rollouts. Enter the 6 million new cases of the last two months.

German Population Data Supports Face-Mask Wearing

On April 6th, the city of Jena became the first in Germany to make mask-wearing mandatory in all public spaces. Notably, to help with PPE shortages, this ordinance allowed the use of scarves or other materials as long as they shielded the face in some way. Soon afterward, the rate of new infections fell to almost zero.

While mask-wearing became compulsory on a federal level between April 20th and April 29th, six regions (including Jena) introduced their own requirements between April 6th and April 25th.

A study published in the first week of June used data from 401 regions in Germany to create a synthetic control group, the unique rolling nature of the regulations allowing them to simulate the specific impact of mask-wearing on the national spread of the disease.

While not all regions showed results quite as drastic as those from Jena, the study concluded that the daily growth rate of COVID-19 cases fell by ~40% due to mandatory mask-wearing policies.

Image Source

While this data cannot be used to isolate the specific reason why mask-wearing slowed the growth rate–e.g. due to source control, personal protection, or just providing a visual cue to maintain social distancing–the results seem quite clear: mask-wearing policies save lives.

Analysis of Global Data

A study updated on June 2nd analyzed coronavirus mortality data from 198 countries up until May 9th. They looked at countries in which masks were worn early on in the outbreak based on tradition as well as those in which the national government mandated mask-wearing by the general public, comparing them to those without such traditions or policies. Based on an average of 23 days from infection onset until death, they only included countries whose policies were in place by April 16th, 2020.

After adjusting for other variables, the researchers determined that:

“In countries with cultural norms or government policies supporting public mask-wearing, per-capita coronavirus mortality increased on average by just 9.0% each week, as compared with 47% each week in remaining countries.”

We could probably stop right here, point made. But more studies, more words.

Studies like this one can never truly isolate a single variable for cause and effect with such a diverse sample pool. Nonetheless, the researchers did a good job of using statistical methods to narrow it down as much as possible, and the effect size is striking.

Let’s now look at a few more individual examples of how universal masking has played out in different countries.


A study found that quarantine and recommended mask usage reduced the peak of the COVID-19 epidemics, on average, by 15% in Sa ̃o Paulo (SP) and almost 25% in Bras ́ılia (DF). The researchers developed a mathematical model of transmission that incorporates infection growth rates across multiple cities and plotted how it was impacted by quarantine and federal mask usage requirements.

As seen above, there is a marked reduction in the transmission rates that begins ~6 days after the use of masks was mandated. While the study wasn’t able to isolate the causal impact of mask use from that of quarantine or other measures, based on the relative timing they conclude that “our empirical analysis supports the view that use of masks seems to be more effective than social isolation.”


In Austria, a mandate to wear masks in shops was announced on March 30, with the expectation that masks would be available by April 1st. This was followed on April 6th with a requirement to wear masks on public transit.

On April 14th, with daily new cases down by almost 70%, Austria became one of the first countries in Europe to begin loosening its lockdown restrictions, allowing all on-essential shops of 400 square meters (~4300 sq ft) or less to reopen.

On May 15th, bars and restaurants were allowed to reopen, so long as they obeyed social distancing guidelines. Only the waiting staff were required to wear masks.

Given the incubation period of the disease, it seems likely that the downturn in cases was not caused solely by the introduction of masks (even if the headlines seem convinced). We’ll have to wait for a proper multivariate analysis to be carried out to isolate some greater cause and effect, but the initial data looks promising.

There are still a few new cases in Austria each day, but the rate has held steady enough that it seems like their current set of policies has managed to get the outbreak fully under control.

Smaller Scale Studies

The studies discussed so far seem to be the only ones currently published or on pre-print servers that evaluate the effect of mask-wearing on a population level. There are, however, several more that use smaller cohorts to continue to add to the body of literature on the positive impact of mask-wearing.


A Chinese study posted on June 4th looked at the relationship between COVID-19 infection and public risk perception, information source, knowledge, attitude, and four non-pharmaceutical interventions(NPI: hand washing, proper coughing habits, social distancing and mask-wearing) during the COVID-19 outbreak in China. In total, 8158 adults were included in the study and 57(0.7%) infected with COVID-19. They found that, amongst these 4 NPIs, mask-wearing had the strongest association with lowered infection rates. Even amongst those practicing the other 3, there was a 16.7% infection rate for those not wearing a mask, compared to a 0.6% amongst those that did.

This study was limited by the small number of infected participants but still shows some very interesting, statistically significant results in favor of mask-wearing as compared to the other interventions.

In China, outside of Hubei, around 80% of infections involved families and close contacts in the household. This is likely an effect of home quarantines since mass gatherings were limited, but it also provides an interesting target for research.

A fascinating study looked at 335 people in 124 families, each with at least one laboratory-confirmed COVID-19. It evaluated the mask-wearing and personal hygiene habits of participants, looking at how they impacted secondary disease transmission rates.

The researchers found that face mask use within the household by the primary case and family contacts before the primary case developed symptoms was 79% effective in reducing transmission, whereas wearing a mask after the onset of the illness symptoms was not significantly effective.

This goes against the common wisdom of “wear a mask outside, take it off when you get home.” It remains unlikely that home mask-wearing will ever take off in western countries, but this study provides strong evidence that it might be a good idea, especially for families of healthcare workers that are regularly exposed to the virus.


A study in Thailand looked at 1,050 asymptomatic participants identified via contact-tracing, asking questions about their habits of wearing masks, washing hands, and social distancing during the contact period.

They found that wearing masks all the time was associated with ~1/4 the risk on infection, whereas only wearing masks some of the time had no significant impact. Interestingly, this was less effective than keeping a >1m distance, similar to limiting the duration of interaction to under 15 minutes, and more effective than hand-washing.

Wearing masks all the time (adjusted odds ratio [aOR] 0.23; 95%CI 0.09-0.60) was 52 associated with a lower risk of COVID-19 infections compared to not wearing masks, while wearing 53 masks sometimes (aOR 0.87; 95%CI 0.41-1.84) was not. Fun fact: sharing a cigarette was associated with a ~350% risk of infection. Big surprise.


A study in Paris tested 1,344 symptomatic HCW for SARS-CoV-2 out of a total of 13,278 employees across 2 healthcare facilities. Preventive measures with the implementation of universal masking, reinforcement of hands hygiene, and social distancing were applied from March 16th in both settings. Seven days later, the epidemic curve in the facilities flattened, although HCW were increasingly exposed to COVID-19 patients and the outbreak was concomitantly at its peak in the region Ile-de-France.

This study was unable to isolate the impact of masks from that of hand washing and social distancing, but it does contribute to the body of evidence for the combination of those three being the best path forward.


Mask-wearing by the general populace seems to reduce the spread of COVID-19 by anywhere from 15-40%. When worn properly and in the right context, mask-wearing might reduce individual risk by up to 80%.

The studies looked at in this article are just the beginning. As COVID-19 continues to progress across the globe, data will become increasingly available and it is likely that the strength of this effect will only be more strongly proven.

The WHO may not think that there exists any “direct evidence” of universal masking being effective. They list many potential disadvantages, such as the increased risk of self-contamination, a false sense of security leading to increased infection rates, and various aspects of personal discomfort.

There has been no published research on a detrimental effect on a population level, and millions of people seem to have managed to deal with the discomfort.

When making any decision, there is always some type of cost-benefit analysis. Here, the cost of a cloth mask is less than $10 and some personal discomfort. The benefit could be seen in millions of saved lives. You do the math.

We are still in the early stages of analyzing all of the worldwide data from the last few months. Maybe some new analysis will be published that entirely changes our perception of the situation. So far the verdict seems clear: masks save lives. Even cloth masks.

In this article we looked at the binary question: to mask, or not to mask. The answer was a clear “to mask.” Unfortunately, this was not enough to answer the question of whether or not I’ll end the summer with a mask-tan. For that, we’ll have to spend some time examining precisely how the disease is transmitted, how masks work, and in what context they need to be worn.

It is possible to maximize protection while minimizing discomfort and getting some sun at the same time.

Stay tuned.