This past week, the recommendation was made for everyone to consider using the use of cloth face masks. In addition, a new idea has popped up that Google searches can predict the spread of the disease. That last one may teach us quite a bit about how to identify COVID-19 early. Let’s dig in.
Everyone Wearing Masks?
After the CDC saying that only sick people should be wearing masks, the organization did an about-face this past week and recommended more widespread mask use:
“CDC recommends wearing cloth face coverings in public settings where other social distancing measures are difficult to maintain (e.g., grocery stores and pharmacies), especially in areas of significant community-based transmission.”
So does this make any scientific sense? Is there new data that shows that this slows the spread of the coronavirus or is this just a move to make us feel good?
First, let’s discuss what we’re really talking about here. Not N-95 respirators nor surgical masks, but masks made by people at home using a sewing machine, scarves, or a t-shirt.
The National Institute for Occupational Safety and Health (NIOSH) performed a study on using clothing such as sweatshirts, t-shirts, and scarfs in healthcare settings (1). The conclusion? That these items could provide only marginal protection against small particles. These are the ones that everyone is so concerned about carrying the virus and hanging in the air.
However, maybe NIOSH is wrong? Is there a clinical study? A randomized controlled trial compared the ability of surgical and cloth masks to prevent healthcare workers from contracting the flu. It found that workers wearing cloth masks were thirteen times more likely to get a respiratory viral infection than those workers wearing medical masks (2).
Some New Research?
A new study was just published in the journal Nature using surgical face masks and patients who had diagnosed respiratory viral infections (NOT the novel coronavirus) (5). The goal here was to see how much viral transmission the masks stopped during actual coughs. The good news is that under real-world conditions, the surgical masks seemed to stop most of the smaller particles (the droplets in this study, at under 5 microns, were bigger than the super small ones everyone is concerned about).
The Common Sense Part
The size of the pores or opening in a high-quality t-shirt are 0.85 mm (30 thread count per inch). Let’s decrease that by 1/3, meaning we’re now using very high-quality fabric at 100 thread count, so the pore size is 0.28 mm or 280 microns. How big are the small cough particles we’re discussing here? They’re less than 1 micron or smaller (3).
So what does that look like?
Here I made the small droplet big enough so we can see it, as to scale it would be a tiny period sized thing.
It would take a thread count at >1,000 to start to stop these droplets. You could try tripling or quadrupling up on 300 thread count fabric, which may work just a little. Surgical masks will stop 20-70% of these small droplets depending on how fast the droplet is moving and how the mask is made (4). In fact, only HEPA filters have pore sizes small enough at 0.3 microns, to stop or filter these small water droplets from coughs.
Hence, in conclusion, the idea that cloth masks will stop the spread of the most problematic small droplets is not founded in any science. This is just a move to keep the natives happy and complacent.
Google Searches that Teach Us About the Virus
There was a really interesting opinion piece by a data scientist diving deep into the type of Google searches being done on COVID-19 symptoms and the extent of the outbreak in that part of the world (6). One of the first symptoms of COVID-19 is the loss of the ability to smell. That google search (“I can’t smell”) correlates reasonably well with U.S. cases confirmed state by state. This idea of using Google searches to track disease outbreaks is not new and has been studied before.
I overlaid the day by day COVID-19 case data from Italy on top of the Italian search data on the loss of smell here (one lags the other by several days as losing the sense of smell is an early symptom):
You can see some fit of the two lines around the double peak in the middle and less at the end of March (this graph is through April 1). Did the Italians recently up their testing capability to ID more cases? That could explain the divergent lines at the end of the graph.
An interesting take-home from this article was that one of the searches that correlated the best with COVID-19 was “my eyes hurt”. So is this a common symptom?
The upshot? Wearing a scarf or t-shirt or even a properly sewn cloth mask is unlikely to do little to protect you or anyone else. However, it may make you or someone else take social distancing more seriously, which may be why we’ve been asked to do this. In the meantime, Google searches could be a reasonable way to track the disease spread without widespread blood tests. Maybe we’ll see a world where big data identifies micro-hot spots in neighborhoods based on Internet searches and then we focus widespread blood testing in these areas?
(1) Rengasamy S, Eimer B, Shaffer RE. Simple respiratory protection—evaluation of the filtration performance of cloth masks and common fabric materials against 20-1000 nm size particles. Ann Occup Hyg 2010 Jun 28;54(7):789-98
(2) MacIntyre CR, Seale H, Dung TC, et al. A cluster randomised trial of cloth masks compared with medical masks in healthcare workers.BMJ Open 2015 Apr 22;5(4):e006577
(3) Zayas, G., Chiang, M.C., Wong, E. et al. Cough aerosol in healthy participants: fundamental knowledge to optimize droplet-spread infectious respiratory disease management. BMC Pulm Med 12, 11 (2012). https://doi.org/10.1186/1471-2466-12-11
(4) Chen CC , Willeke K. Aerosol penetration through surgical masks. Am J Infect Control. 1992 Aug;20(4):177-84.
(5) Leung, N.H.L., Chu, D.K.W., Shiu, E.Y.C. et al. Respiratory virus shedding in exhaled breath and efficacy of face masks. Nat Med (2020). https://doi.org/10.1038/s41591-020-0843-2
(6) Stephens-Davidowitz S. New York Times. https://nyti.ms/2UKc0sZ Accessed 4/6/20.