How Risk-Based Thinking Could Prevent Thousands of Covid-19 Deaths

Design failure mode and effects analysis (DFMEA) and a look at better masks to prevent spread of Covid-19

Published: Tuesday, January 5, 2021 - 13:02

As someone who has helped companies in a wide variety of industries for the last 30 years solve many problems using risk-based thinking, I cannot think of an issue that I have worked on that is more important than preventing the spread of Covid-19. With three high-risk people in my home, I have spent considerable time studying Covid-19 since February 2020. By applying the risk-based thinking techniques I have used, I believe there is a method for saving 100,000 lives before we get the protection the new vaccines are going to provide during the next three or four months.

A look at South Korea

Daily new confirmed cases of Covid-19 per million people: S. Korea vs. U.S. Source: Our World in Data and Johns Hopkins University. Click image to enlarge.

The country that most people consider the gold standard for controlling the Covid-19 virus is South Korea. With its population of 52 million people vs. the U.S. population of 331 million people, South Korea is just under one-sixth the size of the United States. The “Daily new confirmed Covid-19 cases per million people” chart shows the number of new infections per day per million people for the United States and South Korea from Jan. 20, 2020, the date both countries experienced their first infection, through Dec. 4, 2020. The chart shows that on December 4, the United States had 688 new daily infections per million people or about 60 times South Korea’s 11 infections per million people. What is South Korea doing differently?

The published information on South Korea’s success leads one to believe its success is due to testing. My research showed it was something different. According to the “Our World in Data” website, as of December 4, 2020, the United States had performed about 620 total tests per thousand people, or about 10 times South Korea’s 60 tests per thousand people. If testing was the key to success, the United States should be outperforming South Korea, but it is not.

Facial covering usage in the United States has increased dramatically and is now approaching South Korean usage levels. In a National Geographic poll published on October 5, 2020, 92 percent of 2,200 Americans surveyed said they always or sometimes wore a mask when leaving the house. Similar percentages are revealed in a recent CDC study, “COVID-19 Mitigation Behaviors by Age Group—United States, April–June 2020,” shown below.

Source: CDC study “COVID-19 Mitigation Behaviors by Age Group — United States, April–June 2020”

The final analysis showed that the key difference between the two countries was the types of facial coverings being worn by the public of the two countries. The most common facial covering used by the South Korean public is the KF94 respirator. The KF94 respirator has similar performance to the N95 respirator recommended by the FDA and CDC for use by medical personnel. The most common facial covering used by the United States general public is one of the two CDC-recommended cloth facial coverings. I decided to perform design failure mode and effects analysis (DFMEA) on the KF94 and the CDC masks.

DFMEA for KF94 respirator

The first column of the DFMEA is used to capture the design requirements. Every effective Covid-19 facial covering must have two key features to prevent Covid-19 infection of the public by the wearer, and infection of the wearer by the public. The two features are excellent filtration and an excellent seal between the facial covering and the face. Since the published data on South Korea provided evidence that the KF94 was being used to effectively control the Covid-19 virus without a vaccine, its filtration and sealing design requirements were used as design requirements for all the facial covering DFMEAs.

Although Covid-19 aerosol transmission was not originally considered a major source of infection, it is now. At a virtual press conference of the American Association for Aerosol Research, researchers argued the reason that both the World Health Organization (WHO) and the CDC are reluctant to embrace Covid-19 aerosol transmission is in part due to “historical bias,” despite outsized evidence of aerosol transmission. Aerosols commonly contain particles smaller than 1 micron. The filtration design requirement for the KF94 is 94 percent of 0.3 micron particles.

When providing protection for the wearer from infection by the public, an excellent seal to the face is also required to prevent the Covid-19 particles from going around the facial covering. The KF94 respirator design is compliant with fit requirements specified by U.S. Code of Federal Regulations Title 29 Part 1910 Subpart I—Personal Protective Equipment § 1910.134—Respiratory protection (29 CFR 1910.134).

The nose area of the face is the most difficult to seal. A formable insert exists in the nose area of the KF94 respirator to ensure proper fit (see diagram to right). Without this insert, the 29 CFR 1910.134 fit requirement likely could not be met by the KF94 respirator.

Following is the DFMEA for the KF94 respirator for the two key feature design requirements:

The Occurrence column indicates the probability of failure mode occurring due to the failure cause. A “1” in every row, determined by the performance of the design controls, indicates that the KF94 respirator meets the filtration and sealing requirements. This is to be expected because the listed design requirements are two of the actual KF94 respirator design requirements. The Class column is left blank because the probability of the KF94 respirator design failing to meet either of the two design requirements and exposing the wearer or the public to Covid-19 infection is negligible.

DFMEA for CDC masks

The CDC has recommended making your own mask using cotton fabric or other washable, breathable fabric and foloowing a couple different construction methods.

The CDC masks are modeled after the surgical mask. The matrix to the right shows the 0.3 micron filtration capability of the KF94 respirator, medical-grade surgical masks, and 10 of the most common materials used when making the CDC do-it-yourself (DIY) masks. The table is based on experiments conducted by researchers at Missouri University of Science and Technology. The materials used in the CDC masks shown have an average 0.3 micron filtration capability of 25.28 percent, compared with the South Korean KF94 respirator’s 94 percent and medical-grade surgical mask of 78.47 percent. The main difference between the CDC masks and the medical-grade surgical mask is they typically have one-third or less the filtration capability of the surgical mask, depending on the cloth material selected.

None of the CDC masks have a formable insert in the nose area. They also have a rectangular shape since they are modeled after the surgical mask. A surgical mask is designed to protect the patient from infection by the surgeon. It is not designed to protect the surgeon from infection by the patient. Consequently, a good seal to the face is not critical. The picture of the surgical mask on the right shows the typical leak paths in the nose area, chin area, and neck area. These same three leak paths exist with the CDC masks.

Because the CDC masks share the same rectangular shape and filter materials, a single DFMEA will be conducted. Following is the DFMEA for the CDC masks for the two key feature requirements:

The CDC masks would fail the two design controls, resulting in occurrence ratings of 10 for both lines of the DFMEA. A “safety” symbol would be placed in the Class column identifying that the design will result in objectionable risk exposure that could result in death or injury. No commercial company would release the current CDC mask designs for manufacture if they were provided the two design requirements shown in the DFMEA. Their legal exposure if they stated their masks met the requirements could lead to bankruptcy of the company.

CDC acknowledges cloth masks do not protect the mask wearer. The following can be found on the FDA website:

“Cloth face coverings, whether provided by the employer or brought from home by the worker, are not respirators or disposable face masks and do not protect the worker wearing them from exposures. Cloth face coverings are only intended to help contain the wearer’s respiratory droplets from being spread.”

The ‘something is better than nothing’ argument

When people learn the CDC masks do not protect the wearer from infection, they often respond, “Something is better than nothing.” This argument is only true when the only two options are “something” and “nothing.” Those are not the two options facing the American public. The CDC trusts the American public enough to provide instructions on how to make masks that are ineffective. They need to trust the American public enough to provide instructions on how to make an effective mask, which is possible to do for $5 to $7 using commonly available materials.

Features an improved CDC facial covering would likely have

There is little material published on the testing of existing DIY cloth masks. Testing is expensive, and the people creating DIY cloth mask designs are doing it as a public service. Unfortunately, since most DIY mask makers follow the CDC recommendation, most of the DIY masks in the public domain are ineffective. After reviewing more than 100 DIY masks in the public domain, I have found only one that had the four key features found in the KF94 and N95 respirators that I believe make them effective. Fortunately, the CDC has the resources to test any new cloth designs it comes up with.

For multiple reasons, I cannot recommend any DIY cloth mask. By analyzing the KF94 and N95 respirators, I can, however, offer an opinion on the key features that I believe any new CDC-recommended cloth facial covering, and your current cloth facial covering, should have. The key features are:

Nonwoven material filter layer: Woven materials such as cotton are too porous to filter out the small Covid-19 aerosol particles. The KF94 and N95 respirators use a nonwoven material called polypropylene. The picture shows the structure of woven vs. nonwoven materials. The filter layer can be permanent or replaceable.

Multiple sizes with edge profile that conforms with contour of face: An excellent face seal is critical. One size does not fit all.

Formable nose insert: The nose area is the most difficult area to seal. Check out the gap in the nose area of masks you see when you are out in the public. One of my favorite places for viewing ineffective facial coverings is a televised college or NFL football game.

Breathable materials: There is a misconception that materials with excellent filtering capability cannot be breathable.

Washable (not a feature of N95 and KF94 respirators): People cannot afford to replace their masks.

If you would like more information on these key features, check out the YouTube video, “DIY Mask Basics.”

Action is required during vaccine implementation

On Dec. 3, 2020, the Institute for Health Metrics and Evaluation (IHME) at the University of Washington’s School of Medicine released an article titled, “Despite vaccinations, Covid-19 death toll likely to increase into Aprii.” In the article they state: “The forecasts now include scenarios of rapid vaccine rollout, expected rollout, and no vaccine. A rapid vaccine rollout is forecast to reduce the death toll to 528,000 in the U.S. The forecasts show that if mask-wearing increased to 95%, combined with expected vaccine rollout, approximately 66,000 lives could be saved, compared to a vaccine rollout scenario with current mask-wearing levels remaining the same. Even with a vaccine, if states do not act to bring current surges under control, the death toll could reach 770,000 by April 1.”

Due to the high usage of highly ineffective masks that already exists, I am not sure getting the usage to 95 percent of the same ineffective masks will result in a savings of 66,000 lives. Based on the fact that the U.S. death rate due to Covid-19 of 0.93/1,000 is 93 times greater than South Korea’s death rate of 0.01/1,000, and that 220,000 additional deaths are predicted by the IHME by April under the best circumstances, it is probably a conservative estimate that an effective mask that both prevents the wearer from getting infected by the public and the public from being infected by the wearer could prevent 100,000 or more Covid-19 deaths.

About The Author

Richard Harpster

Richard Harpster is president of Harpco Systems, which he founded in 1987. Harpco Systems specializes in providing software, training, and consulting for risked-based product lifecycle management (RBPLM). During the past 30 years, Harpster has helped hundreds of companies implement improved risk-based design and manufacturing systems in a wide variety of industries. He is a recognized expert in the application of FMEAs and has invented several new concepts, including the linking of design FMEAs to process FMEAs in 1990, which became an automotive industry standard 18 years later. His latest inventions in the field of RBPLM include Requirements Risk Assessment (RRA), Usage Risk Assessment (URA), Multiple Integrated Cause Analysis (MICA), and Rapid Integrated Problem Solving (RIPS). He has published several papers on the topic of RBPLM.