• 
• 

Three Part Question

Do [homemade face masks] work as a preventive measure in [the general population] for [respiratory virus transmission]?

Clinical Scenario

Since the outbreak of COVID-19, we are facing an imminent shortage of surgical facemasks and respirators to protect patients and healthcare workers. Inspired by the widespread use of cloth masks in Asia, alternatives out of cotton and old bras are on the rise in Belgium. This raises the question whether these homemade facemasks are effective in the prevention of virus
transmission. And if so, which materials are most suitable.

Search Strategy

Pubmed, Embase, Cochrane and the specific COVID-19 database LitCovid, conception until 20/03/2020
"Masks"[Mesh], “Viruses"[Mesh], "Coronavirus"[Mesh], infection, "Respiratory Protective Devices"[Mesh], "Developing Countries"[Mesh], "Masks"[Mesh], "Influenza, Human"[Mesh], 'surgical mask'/exp, 'mask'/exp, 'coronavirinae'/exp, 'droplet infection'/exp, 'cloth mask', virus, 'cotton'/exp, 'respiratory tract infection'/exp, 'cotton'/exp, 'face mask'/exp, 'coronavirinae'/exp, 'face mask'/exp, 'influenza virus'/exp,
Mask, droplet infection, coronavirinae, “homemade protection masks “, "Pandemics"[Mesh], "Masks/statistics and numerical data"[Mesh], "Respiratory Protective Devices"[Mesh], "Cotton Fiber"[Mesh], cotton,
nanofabric, microfiber, muslin, influenza, respiratory infection, “cloth masks”

Search Outcome

848 articles

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
MacIntyre CR 2015	- 1607 healthcare workers (HCW) > 18 y/o (nurses and doctors) - 14 secundary/ tertiairy - Hospitals in Hanoi	Randomised controlled trial - Cluster randomisation by using Epi info V.6 for allocation of 74 wards (emergency, infectious/respiratory disease, intensive care and paediatrics) containing 1607 HCW. - Daily follow-up - Labo testing and results were blinded. Clinical end points unblinded. - Compliance monitored by a self- reporting mechanism. - ITT analysis, yet no withdrawals.	Rates of all infection outcomes highest in the cloth mask arm	ILI (relative risk (RR) =13.00, 95% CI 1.69 to 100.07)	- No allocation bias - No withdrawal bias - No attrition bias - No reporting bias - Ascertainment bias due to unblinded clinical end points, since a facemask is a visible intervention. - Lack of a no-mask control group: high rate of mask use in the controls through standard practice mask use impedes interpretation. - Recall bias: compliance was measured through a self-reporting mechanism. - Quality of cloth masks varies, not generalizable.
			ILI and confirmed viral infections	ILI (RR = 6.64; 95% CI 1.45 to 28.65) and confirmed viral infections (RR = 1.72; 95% CI 1.01 to 2.94) significantly higher in cloth mask group
			Penetration	cloth masks 97%, medical masks 44%
MacIntyre CR 2015 Vietnam	People in community setting and healthcare setting	Review (not SR) of 9 cluster RCT’s, of which 3 concerning cloth facemasks: - Larson et al. 2009 (open label) (3) - Cowling et al. 2010 (single blind) (4) - Suess et al. 2012 (single blind) (5)	Community setting	people in high risk settings could benefit from facemasks.	- Confounding effect due to pooled analysis with medical masks or simultaneous use of hand sanitizer. - Low compliance - Self-reporting bias - Performance bias: use of masks in control group. - Suboptimal fit of the masks
			Healthcare setting	only medical masks and respirators offer enough protection.
			use of cloth masks	cloth masks only when no other option
			Compliance	Compliance is a determinant of protection.
Davies 2013	- Healthy volunteers - 12 men, 9 women ( 20-44 y/o)	In-vitro and in-vivo observational study	filter efficiency	Household materials have a filter efficiency of 50-89% against 20nm particles	- Not enough power - Higher than realistic compliance - Suboptimal fit of the mask - Population sampling error possible
			forming of bacterial colonies	No difference in forming of bacterial colonies
Van der Sande 2008	- 1st exp: 28 adults and 11 children (5- 11y/o) - 2nd exp: 22 adults (10 men en 12 women) - 3rd exp: artificial in- vitro	In-vitro and in-vivo observational study	protection	Protection mainly depends on type of mask.	- Not enough power - Higher than realistic compliance - No data on couching - Population sampling error possible
			degree of protection	Tea cloth masks still offer some degree of protection
			decay of protection	Protection decreases over time
Rengasamy 2010 USA	not applicable	In-vitro observational study	penetration	Cloth 74-90%. - Cotton/polyester T- shirt 40-% depending on composition. - Cotton T-shirt > 85%. - Towel 60-66%. - Scarf 73-89%.	- Not enough power - Use of nano- particles, not viruses - Sampling error of fabrics possible
Shakya 2017 USA	not applicable	In-vitro observational study	penetration of regular cloth masks	60-80% by particles < 100nm	- Not enough power - Use of nano- particles, not viruses - Sampling error of fabrics possible
Golanski 2009 France	not applicable	In-vitro observational study	penetrance	Cotton has a penetrance of 27% at low velocity particle flow. Industrial fabrics are significantly less penetrable.	- Not enough power - Use of nano- particles, not viruses - Sampling error of fabrics possible - Velocity of particle flow is not representative for airflow when breathing or coughing.

Comment(s)

None of the in-vivo studies recommend the use of cloth facemasks as they offered poor protection against viruses. In-vitro studies showed that household materials such as cotton and cotton-polyester blend are mostly permeable for virus-sized particles. The level of protection provided by homemade facemasks is mainly determined by the used material, compliance and correctness of fit. The majority of studies caution against the use of cloth masks made of fabrics such as cotton or cotton-polyester blend, for the prevention of virus transmission, especially for health care professionals. Infection rates have shown to be much higher in cloth masks compared to disposable medical masks. Hence, most of these masks are over 50% permeable to nanometer-range particles, and therefore provide very little protection to the COVID-19 droplet particles which range from 50 to 200 nm. Heavy materials such as towels and tea cloths perform slightly better compared to T-shirts, yet have to compromise on respiratory comfort, which reduces the compliancy of use. Furthermore, due to a poor fit, leakage of viral particles reduces their protective effect. However, if respirators or surgical masks are not available in a pandemic, cloth masks may be used on the principle that ‘something is better than nothing’. Nevertheless, this strategy is highly discouraged for healthcare workers, who are at a higher risk of exposure, given that a cloth mask will not provide sufficient protection. Scientific data are sparse. Other than one RCT, no studies have been conducted concerning the efficacy of cloth masks, due to the introduction of disposable medical masks and respirators. Consequently, it is difficult to determine whether these homemade cloth masks offer any clinical protection. Further profound research and international guidelines are needed.

Clinical Bottom Line

Homemade face masks provide insufficient protection against viruses. Therefore, their use is cautioned in any health care setting.

References

1. MacIntyre CR A cluster randomised trial of cloth masks compared with medical masks in healthcare workers BMJ Open 2015
2. MacIntyre CR Facemasks for the prevention of infection in healthcare and community settings BMJ 2015
3. Davies Testing the Efficacy of Homemade Masks: Would They Protect in an Influenza Pandemic? Disaster Medicine and Public Health Preparedness Disaster Medicine and Public Health Preparedness 2013
4. Van der Sande Professional and Home-Made Face Masks Reduce Exposure to Respiratory Infections among the General Population. PLoS One 2008
5. Rengasamy Simple Respiratory Protection—Evaluation of the Filtration Performance of Cloth Masks and Common Fabric Materials Against 20–1000 nm Size Particles The Annals of Occupational Hygiene 2010
6. Shakya Evaluating the efficacy of cloth facemasks in reducing particulate matter exposure J Expo Sci Environ Epidemiol 2017
7. Golanski Experimental evaluation of personal protection devices against graphite nanoaerosols: fibrous filter media, masks, protective clothing, and gloves. Human & Experimental Toxicology 2009