We are all faced with the overwhelming amount of information on social media and the news regarding COVID. By far, the most common debate I see on social media is whether we all should be wearing face masks. A lot of people quote newspaper articles and incidental evidence they read on other posts online. As a scientist who has published over 30 peer reviewed articles, as well as a reviewer for international medical Journals, I thought It would be prudent to write a small review of the current evidence. I think it is important to get an unbiased and evidence based opinion on this. I pride myself of being able to look at things with a critical and unbiased point of view.
The biggest mistake we can all make is to do something based on opinions that have been propagated on social media. It is important to understand, things that make sense are not always best when it comes to medicine and science. A good example of this is anti-cough medicine (Cough suppressant). There is no evidence that cough medicine for kids or adults works. If you do a study comparing it to placebo, the placebo works just as well, but yet admittedly, even I purchase some good ol’ NyQuil with cough suppressant when I have the flu or a cold.
There is also a difference in science between absence of evidence and evidence of absence. Meaning, if we have data that says a treatment isn’t good, than we should not use that treatment. However, if we do not have data at all, then we shouldn’t assume the treatment doesn’t work, but rather, investigate it. This is the current challenge right now. Do we have enough data to support the benefit of facemasks for Covid and if not, what data do we have to make recommendations?
To begin, evidence that face masks can provide effective protection against respiratory infections in the community is almost non-existent, particularly in Canada. We must make assumptions that data taken from hospital studies as well as research in countries with a greater population density (ie China, South Korea) also apply here in North America. This is another important point to remember. In science, what may work in one situation, may not work when applied to another situation. For example, studies that assess the benefit of masks in a hospital setting occur in a high risk environment with a high density of people (many people in a small area). This is very different than walking down the street with outside air and low density of people who are low risk. These are two VERY different situations and thus applying the results of one situation to another may not be ideal.
Second, most of the reports we are getting now about recommendations for facemasks are based on ‘expert opinion.’ In science (and medicine) we grade research by assigning a “Levels of Evidence.” What this means, is that some types of research studies are better than others. There are 7 levels of evidence. The BEST is a Meta-analysis of Randomized Controlled trials. A Meta-analysis allows you to combine a bunch of high quality studies and therefore get a very detailed and accurate assessment. One Randomized controlled trial may have 80 patients, but if you combine 10 studies you may now have 800 patients. The next best type of study is a double blind, randomized controlled trial. What this means, is that you take people and randomly assign them to two groups. Then, one group is given a treatment and the other group is given a placebo (something that doesn’t work). The key is, that no one knows if they got the treatment or the placebo. The ‘double blind’ nature means that, the person who is giving the patient the treatment or placebo and collecting the data also doesn’t know what treatment the patient gets, so they can not be biased. You can imagine, there is no way to do this for facemasks. We can not give someone a placebo (no facemask) and another group surgical masks and expect they not to know what they are wearing! So, if you wanted to test the utility of a facemask in preventing illness in a hospital, usually the studies compare different types of masks as opposed to comparing one mask to no masks.
The LOWEST quality evidence is ‘Expert Opinion.” Yes, experts can be wrong. If expert opinion is based on the data from a meta-analysis than that is high quality evidence, if expert opinion is based on their personal observations, that is low quality evidence.
Background on COVID
Covid is a type of virus in the Coronavirus Family of viruses. It is considered a respiratory virus that spreads via droplets. The Covid virus is not all alone floating through the air, it is attached to saliva/mucus and other things from our mouth and nose when we cough/sneeze. We’ve all been hit with a sneeze before, that moisture is what Covid attaches to. So, droplet doesn’t imply it’s not in the air, this is a misconception by many. Once the droplets are in the air they can then land and stay infectious on whatever was sneezed on.
The Covid virus is also VERY small, it is 0.1microns in size. Smoke particles are 10x larger at 1 micron while dust is about 100x larger at 10 microns. The actual droplets are much larger than this, over 25-100 microns. It is controversial if Covid is travelling on its own without droplets. If it was, then N95 masks would be the only way to prevent the spread of the actual virus.
Types of Facemasks
There has been a lot of information spreading on social media about masks and the effectiveness of different masks. I have yet to see any of these images, graphics or posts demonstrate their sources and if this is ‘expert’ opinion or evidence based on research. I’ll do my best to provide the details (as appropriate) and ONLY include published, peer reviewed research studies.
- Respiratory Mask (an N95) – these are airtight and filter out small particles and any biological aerosolized particle. To use these, you need a special ‘fit test’ to ensure that they are airtight on your face and particles don’t come in around your nose. These are the ‘best’ masks for filtration. When worn properly, they achieve about 95% Filtration of particles over 0.1micron. If the droplet is 0.75 microns or larger, they are 99.9% effective, which is likely the case for droplets of COVID19.
- Surgical/Procedure Masks – I wear these during surgery and when doing a procedure. They are meant as a barrier to droplets, like blood, snot and other forms of goo. They are NOT airtight no matter how tight you tie them. Some people believe they are ineffective in preventing airborne infections.
- Cloth Masks – These would be homemade/consumer grade masks that are not used in Hospitals. This have been used historically and are common in some countries. Cloth masks can be made from different materials which also affect their usefulness.
Comparing Surgical Masks to N95 Masks
I was able to identify 3 Meta-analysis which attempted to determine if N95 Masks were better than standard procedure/Surgical Masks for preventing infections
Study 1: The first was conducted in 2017 assessing the difference between N95 Respirators and Surgical Masks. The study wanted to determine how each masks protected respiratory infections (colds, flu and things like Corona Virus) in health care workers (Clin Infect Dis, 2017;65(11)). This study ONLY looked at hospitals, which are high risk areas. This was a very good, well performed meta-analysis.
The study identified 4 Randomized Controlled trials that assessed the difference between N95 masks and surgical masks in the health care setting. In total, 1989 people wore surgical masks and 2464 wore N95 masks. Two of these studies were conducted in North America and 2 in China. ALL of these studies also tested for Corona Virus using PCR (which is a very accurate method to detect the virus). The authors used an outcome data point called “Odds Ratio.” This means, that if you compare N95 to Surgical Masks, what are the Odds one mask is better. When combining all studies, there was no difference between masks (Odds Ratio around 1.0) for Influenza (the flu) and general clinical respiratory illness (other viruses).
The study found that that N95 Masks were slightly superior to surgical masks when preventing viral infections (like Covid). The study did not report if masks where better than NO masks. So, this study doesn’t really tell us much apart from, N95 masks and surgical masks are both effective for Viruses in a hospital setting.
Study 2: The second meta-analysis (Smith et al, 2016, CMAJ; 188(8)), was published in the CMAJ (Canadian Journal). The difference with this study, is that it included more than randomized controlled trials. The subjects in the included studies were health care workers. They did not include any data on community spread of the flu or viruses. The study again assessed whether N95 masks were superior to surgical masks. The findings were similar to the previous study, there was no difference between N95 masks and Surgical masks when used to prevent respiratory virus infections, they were both effective.
Study 3: The third was conducted in 2020 (Long et al, 2020. J Evid Based Med: 1-9) and was primarily concerned with Influenza, which is also a virus. Similar to the other two studies, there was no difference between N95 masks and surgical masks but again, both masks were very effective.
SUMMARY – Both N95 masks and Surgical Masks are effective in preventing Viral Infections in the Hospital.
Comparing Surgical/N95 Masks to Paper/Cloth Masks
Given the shortage of hospital masks, many countries are looking for other options to protect their population. One option is the use of homemade or consumer grade masks, cloth masks.
STUDY 1: A randomized trial was conducted in Australia in 2014 comparing cloth masks to standard hospital masks (MacIntyre et al. 2015. BMJ). 15 hospitals were included with a total of 1607 participants. The participants were randomized to wear either Surgical Masks or Cloth Masks. All participants were health care workers in the hospitals who worked in high risk areas (Pediatrics, Emergency Room, ICU and Infectious disease). 580 people wore Surgical Masks at all times during their shift, 569 wore Cloth Masks at all times and 458 wore surgical masks when they believed it was required. The study assessed three outcomes, (1) Getting a clinical respiratory illness (Cold) based on symptoms (2) Getting Influenza (the Flu) (3) laboratory confirmed viral illness including Corona virus (SARSCoV). Those who wore cloth masks washed them at the end of every shift. The workers were followed daily for 4 weeks.
The Results of this study indicated that surgical masks resulted in significantly fewer infections than cloth masks. For Influenza, wearing a cloth mask resulted in an increased the risk of developing an infection compared to a surgical mask by 13.2 times. When assessing general respiratory viral illness, wearing a cloth mask increased the risk of general infection compared to medical mask by 1.57. Interestingly, the study looked at some other variables in what is called a “multivariate analysis.” This means, they looked at mask type, Vaccination and Handwashing all together and determined which had the biggest effect on reducing or increasing the risk of infection COMPARED to using a medical mask. For Influenza, the use of Cloth Masks resulted in a 13 times risk of getting an infection. Adding handwashing to the use of surgical masks significantly reduced the risk of getting an infection even more.
What this study tells you, is that cloth masks are inferior to surgical masks. HOWEVER, what it doesn’t tell you, is if cloth masks are better than no mask. The study did not look at this so we do not know how cloth masks would compare to no mask. Also, it highlights the important of handwashing.
STUDY 2: During the SARS pandemic in the early 2000’s, a number of studies looked into the use of a paper mask. SARS is similar to COVID and thus, data likely has some relevance to our current pandemic. Seto (2003) published a case control study. This means, the authors identified patients with SARS and without SARS who worked together in 5 Hong Kong Hospitals. The goal was to determine what may have contributed to those who didn’t get infected. This was a small study with only 13 health care workers infected while 241 were not infected.
The results of this study indicated that there were no infections in people if they wore either a surgical mask or N95 mask. Two people out of 171 were infected if they wore a paper mask, however, this is a very small number and not significant from a statistical point of view. It is important to note, that this was a small, retrospective study which means the data is not high quality. It is very difficult to make a determination based on this study. From a scientific point of view, this study does not provide much evidence other than suggesting facemask are effective in general. The study however, also noted that handwashing was important.
STUDY 3: Another case control study in 2012 (Zhang et al, 2012, Influenza) assessed hospital workers during the H1N1 Crisis. The results of this study demonstrated that Facemask use did not have an effect on virus spread. The only factor that significantly reduced the risk of infection in this study was receiving a vaccination for H1N1. This study highlights the critical importance of developing a vaccine.
STUDY 4: Perhaps the most interesting study assessing cloth masks was conducted in 2010 by Rengasamy et al (2011). This was a highly controlled and very well conducted study. This study used a highly controlled environment to determine the transmission of different size particles and different speeds of travel of the particles for 5 types of fabric, compared to N95 masks. The speed of the particles generally corresponds to sneezing and coughing. The study assessed T-shirt Cotton, Cotton Towel, Head Scarf, Sweatshirt and brand name cloth mask advertised for this purpose. The study tested a range of particle sizes from 0.2microns to 1 micron. There were two types of particles used for testing, Polydisperse and monodisperse. Polydisperse Particles is when all sizes of particles from 0.1 to 1 microns us used for testing. This is likely closer to what you would expect in real life with a sneeze. Mono means only ONE particle size was tested at a time.
For Polydisperse testing, N95 masks blocked 99% of all particles at the two speeds tested. The faster the speed of the particles, the less effective the N95 mask worked, but it was still filtering over 95% of particles in all cases.
Sweatshirts and t-shirts resulted in less than 20% filtering of the particles, thus letting 80% through even at low speeds. At high speeds, only 10% of the particles were filtered. Interestingly, Hanes brand t-shirts blocked 50% of particles at low speed. Thick towels blocked about 30-40% of particles. Head scarfs blocked 20-25% of particles at both speeds.
To summarize this paper, cloth masks resulted in anywhere between 10-90% penetration of particles when using the polydisperse method compared to less than 1% using an N95 mask. This does not say cloth masks are bad, but they are unpredictable.
SUMMARY ON CLOTH MASKS
Right now, it is clear that surgical masks are much better than cloth masks. Wearing a cloth mask when access to a surgical mask is available will increase your risk of infection. However, we have a surgical mask shortage, so is a cloth mask better than nothing? The issue remains, we don’t know. There are no good studies in the general, non-hospital population. Based on the study above, cloth masks filter at least some particles, so it is likely they do something, but certainly not as much as a surgical or medical based mask.
APPLYING THE SCIENCE TO REAL LIFE SITUATIONS
Now that we have some background, lets look at more real life situations and try to determine what we all should be doing.
Can infected individuals reduce the risk of spread wearing a facemask?
Yes, facemasks are designed for this purpose. If you have a cough, if you are sneezing, if you have been around a COVID19 positive person or are considered at risk (recent travel), you SHOULD WEAR A MASK to PROTECT OTHERS. This is why we wear masks and why even in my clinic, we ask patients with a cold to wear a mask in the waiting room. Wearing a mask and then practicing proper hand hygiene after taking the mask off is the best way to prevent spreading to someone else. There is plenty of data in this area and this should be an answer we all agree on.
Based on This, The Following People Should Wear Masks To Protect Others
- People with Symptoms of Covid (Cough/Sneeze/Runny Nose/Fever)
- People with COVID
- Contact with COVID+ Individual
- Recent Travel to High Risk Area
- Working in high risk area with exposure to COVID positive individuals
Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in Wuhan, China. JAMA. 2020 Feb 7.
A familial cluster of pneumonia associated with the 2019 novel coronavirus indicating person-to-person transmission: a study of a family cluster. Lancet. 2020 Feb 15;395(10223):514-523.
Can Uninfected People Reduce the risk of infection by Wearing Facemasks?
This is the #1 controversy right now. In this situation, what we are really asking is, does wearing a mask prevent the virus from infected people getting into my mouth/nose/eyes (remember, it can transmit via the eyes like most respiratory viruses). When looking at the data, it is important to break the information down into two groups of people, High Risk and Low Risk.
HIGH RISK INDIVIDUALS – This would include people such as hospital workers or others who have close contact with Covid + People. As you have seen from all prior studies, most of the data looks at high risk individuals, hospital workers. We can only surmise based on this data and apply it to a general population outside the hospital.
Two meta-analysis have demonstrated that in health care workers in high risk environments, masks can reduce the risk of infection by 40-60% (including SARS), this is even after taking confounding variables like handwashing into account. This includes both N95 and Surgical Masks.
LOW RISK INDIVIDUALS – These would represent most people in Canada who do not work in Health Care or high risk jobs. The important feature here is the density of the low risk population. What I mean is, if you have 100 low risk people spread out over 100 km, that is very low population density, it is harder to come in close contact with other people. If you have 100 people in the same department store, that is high density and it is easy to get close to others. Studies that have assessed low risk individuals have been conducted in HIGH density areas like Hong Kong and large Chinese cities. So whether or not we can apply this to Canada is debatable.
A study in Hong Kong and Beijing during the 2003 SARS crisis demonstrated that using masks in high density public spaces reduced the risk of infection by 60%. BUT these were observational, low quality studies. The evidence was poor, but in their study, they still noticed the difference. It is hard to know if this can apply to Canada.
Can Widespread Use of Facemasks Control a Pandemic?
This is also a crucial question and one that people are arguing over. Let’s do a thought experiment. Let’s assume, based on all the previous research, that facemasks may help reduce the risk of COVID infections. Cloth masks may help, even if only a small amount. But, will this help slow the rate of transmission? By slowing the rate, will this control the Pandemic or just prolong and delay it? Will it help flatten the curve? Using surgical masks may do the same, but only much better. I think this is a possibility.
Another argument for wearing masks during a pandemic is that it increases the Social Norm form wearing masks. This can also increase the commitment to infection control in society and thus Increase overall hygiene methods. In North America, it’s fair to say many of us find it odd to see people walking around wearing masks, it is not part of our society. But may they be onto something?
One argument for adopting a more general use of masks is that if more people wear masks, more may accept it as a social norm and then are more likely to wear a mask. Those who are sick will no longer feel a stigma about wearing them in public, thus protecting us all. It sounds silly, but we all know its true. The effect of stigma can be powerful.
The problem is, again, there is no real data into this so we don’t know the answer for this.
What are the RISKS of Wearing Masks?
I’ve been thinking about this a lot. Many health organizations as well as news reports argue wearing masks can cause a false sense of security or make you touch your face more. I always get annoyed when my medical colleagues lecture people on following evidence but then make recommendations based on no evidence. It happens all the time in medicine. So, here are a few risks and what I found out.
Shortage of masks for High Risk people. This is a Big problem. We are running out of masks for high risk individuals, particularly health care workers. If health care workers do not have them, more of them will get sick and less physicians, nurses and hospital staff will be available to care for you. This is not to be underestimated. Health care is high risk and is the area in which there is NO debate. Masks MUST be worn. This is a big issue and why we need to know when to wear masks, particularly hospital grade masks.
Wearing Masks Increases touching your face. I hear this all the time, but is there evidence? There has been one study that studied this. Similar to other studies, it was conducted in a hospital with nurses as subjects (Rebmann et al, 2013. Am J Infec Control, 41:1219-23). This also only looked at N95 masks, which I can tell you from firsthand experience, are very uncomfortable.
In a 12 hour shift, nurses (when wearing the mask) touched their face an average of 5.8 times, touched the mask an average of 14.6 times and re-adjusted the mask 6.6 times. Each time they did this, they put themselves at risk for spreading a virus. Another important point is comfort and safety. Over the 12 hours, there were higher reports of ‘light headedness’ when wearing the masks. Also, the CO2 (Carbon dioxide) in the blood raised from 32.4 to 41 by the end of the shift. Remember, this is a 12 hour shift, so general use by the public would not likely notice these changes in CO2.
The unresolved question is, if you touch your face with a mask, does that increase your risk of infection? As we have seen from all the previous studies, using a mask significantly reduces your risk of infection. But, does the risk of getting an infection from touching your face when wearing a mask outweigh the benefit from wearing it? There is no evidence for this and personally, I now think the benefits of wearing a mask outweighs the risk of touching your face, when in a high-density high-risk environment. It remains unclear if this is also the case in low risk, low density areas.
But, for those who are concerned about spreading the disease when people touch their face while wearing a mask, I ask you this: Are you willing to put people at risk by limiting the use of a proven technology due to an assumption that is not-proven? Its not an easy question.
SUMMARY and RECOMMENDATIONS
My recommendations would be the following:
Wear an N95 Mask or Surgical Mask if you meet any of the following
- Within a High Density area with poor ventilation (Public Transit)
- Within High risk area (hospitals/Health Care workers, patients in the hospital)
- You Have Covid
- You have been exposed to Covid + Patient and are isolating with Others in the house or going out in public
- You have travelled and at risk of covid
You should NOT wear a SURGICAL mask, but can consider a cloth mask if
- in low risk environments, ie Home, driving, walking down a calm city street
- In a public area that is well ventilated (outdoors)
- a low density area (Shopping with few people)
There are still two questions I struggle with:
- Should all health care professionals always wear a mask in the hospital? Based on my research thus far, I think this should occur. As mentioned above, most authorities recommend those who have been exposure to COVID self-isolate and wear a mask. I think it is fair to deduce that most, if not all, health care workers are in close proximity to COVID patients. Therefore, they are part of this high risk group. It may not be reasonable given the shortages of masks, but I do think, based on research that anyone working in a high risk area should wear a surgical mask at all times.2) A second question, should all patients in the hospital wearing a mask , regardless if they are symptomatic and regardless of the reason for being in the hospital? Again, I think the answer is yes. This is a high risk, high density area and in my opinion, they are no different than the hospital staff. In fact, they are likely at higher risk. All patients and visitors in the hospital should always be wearing a mask. It’s my opinion that once you walk into the hospital, you should put on a mask and wash your hands, no exceptions.If we get to the point where no surgical masks are available, should we wear cloth masks? Again, I believe yes. It is a high risk, high density area most definitely. In low risk, low density areas, I think wearing a cloth mask is up to the individual. It likely would not be required, but I do not think there is evidence of harm and so, people should have the right if they so choose. Even if we can reduce the spread by 10- 40% using a cloth mask, that is better than 0%.
What Are Countries Doing Now?
A lot of what I read is people indicating the effectiveness of masks based on the results of countries like South Korea and China. But, what are these countries actually doing? Its important to remember that these countries have also pursued some of the most rigorous testing and isolation protocols in the world. So, is it the masks or the other interventions that are reducing the numbers?
The Lancet (one of the top Medical Journals in the World) just published some information on this. Specifically, what countries are recommending in terms of masks. Let me summarize.
If Healthy, only wear a mask if taking care of a person with Positive or suspected COVID
- Moderate Risk People (People who work in high population density jobs or live/work with someone who some one guaranteed or work in a field related to COVID
- Wear Disposable Surgical Mask
- Low Risk People (In high population area, supermarket/shopping or health clinic)
- Wear Disposable mask for medical use
- Very Low Risk – (Most of Us In Ottawa Would be this) Well ventilated areas, home, driving, low population
- NO Mask
Hong Kong Recommendations
- Use when you are ill to protect others
- Use mask in crowded place or public transport
- MUST Practice Good hand hygiene pre/Post Mask
- Wear Mask if you have respiratory symptoms (cough/Runny nose)
- Masks not efficient unless you are in a closed space with high density of people
- Masks not recommended otherwise
- Only recommended in High Risk places like hospital or crowded area
- No evidence of usefullness of masks